USF Libraries
USF Digital Collections

Linkage-based prosthetic fingertips

MISSING IMAGE

Material Information

Title:
Linkage-based prosthetic fingertips analysis and testing
Physical Description:
Book
Language:
English
Creator:
Ramirez, Issa A
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Linkage-based fingertips
Stability analysis
SHAP test
Gripper
Hook
Dissertations, Academic -- Mechanical Engineering -- Masters -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: This thesis consists of the research on linkage-based fingertips for prosthetic hands. These fingertips consists of small polycentric mechanisms attached to what would be the pulp in normal anatomical fingers. These mechanisms allow the prosthetic hand to conform to the shape of objects during grasp. The goal of these prosthetic fingertips is to maximize the functionality of the hand while minimizing the number of inputs that the user has to control. The stability of the fingertip mechanisms is analyzed using the principle of virtual work. From this analysis we are able to show that the fingertip mechanism is stable for a large range of rotation of the link and for a large range of directions on which the force is applied, and that the mechanism is indifferent to the magnitude of the force applied to it (assuming that the force does not damage/deform the mechanism). To assess if the four-bar mechanisms (fingertips) improve the grasping capabilities in robotics and prosthetics, tests were performed on prosthetic hands and robot grippers with and without the fingertips. Comparisons were made using the Southampton Hand Assessment Procedure (SHAP) protocol, which tests the differences and measures the functionality of particular types of grasp, such as power, spherical, lateral, tripod, tip and extension. In the human testing, the overall Index of Functionality (IOF) of the Hosmer hook is 66.65 and 66.21 for the hook with the fingertips. The hook with the fingertips had a better IOF in the spherical and power prehensile pattern. When the IOF is calculated for the tasks that the fingertips were used, in 10 of 11 of the tasks, the IOF is higher than using the Hosmer hook. In the robotic gripper testing, the Index of Functionality was not be calculated because the time to perform the tasks depended more on the robotic control system than on the physical characteristics of the gripper.
Thesis:
Thesis (M.S.)--University of South Florida, 2007.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Issa A. Ramirez.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 107 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001935133
oclc - 225864686
usfldc doi - E14-SFE0002273
usfldc handle - e14.2273
System ID:
SFS0026591:00001


This item is only available as the following downloads:


Full Text

PAGE 1

i Linkage based Prosthetic Fingertips: Analysis and Testing by Issa A. Ramirez A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Mechanical Engineering Department of Mechanical Engineering College of Enginee ring University of South Florida Major Professor: Craig P. Lusk Ph.D. Nathan Crane, Ph.D. Rajiv Dubey, Ph.D. Date of Approval: November 6, 2007 Keywords: linkage based fing ertips, stability analysis, SHAP test, gripper, hook Copyright 2007 Issa A. Ramirez

PAGE 2

i Table of Contents List of Tables i v List of Figures vii Abstract x Chapter 1: Introduction 1 1.1 Motivation 1 1.2 Scope 2 1.3 Thesis o verview 2 Chapter 2: Background 3 2.1 Epidemiology 3 2.2 Prosthesis 3 2.3 Terminal devices 4 2.4 Commercially available terminal devices 5 2.5 Users priorities 9 2.6 Design of terminal devices criteria 10 2.7 Assistive robot manipulators 10 2.8 Underactuation 11 Chapter 3: Kinemati c Analysis 13 3.1 Analysis of four bar mechanism 13 3.2 Principle of virtual work 14 3.3 Crossed four bar mechanism 15 3.4 Stiff hinged mechanism 17 3.5 Stability analysis 17

PAGE 3

ii Chapter 4: Design 23 4.1 Fingertips 23 4.2 Hooks 24 4.2.1 Manufacture 26 4.3 Robot gripper paddles 26 Chapter 5: The SHAP T est 28 5.1 Prehensile patterns 28 5.2 Southampton hand assessment procedure 30 5.3 SHAP Test 33 5.3.1 Abstract objects 33 5.3.2 Activities of daily living 34 5.4 Index of functionality 35 5.5 Human s ubjects 37 5.6 Robot gripper 38 Chapter 6: Results 40 6.1 Normative data 40 6.2 Hook testing 40 6.3 Observations 42 6.4 Data analysis 43 6.4 Index of functionality 54 6.5 Robotic gripper t esting 57 Chapter 7: Conclusion and Recommendations 59 7.1 Conclusion 59 7.2 Recommendations for futu re work 59 References 61 Appendices 65 Appendix A:Instantaneous stabil ity of the crossed four bar mechanism 66 Appendix B: SHAP Test 72

PAGE 4

iii B.1 Setting up the assessment 72 B.2 Procedural notes 72 B.3 Abstract ob jects 73 B.4 Activities of daily l iving 75 Appendix C: Human testing data 82 Appendix D: Mean values of each subjec t separated by prehensile pattern 102

PAGE 5

iv List of Tables Table 2.1: Functionality and weight of body powered terminal devices commercially available 7 Table 2. 2: Trans radial body powered items for improvement [12] 9 Table 2.3: Design criteria of some terminal devices in research 10 Tab le 3.1: Geometric p arameters 17 Table 5.1: 32 Table 6.1: Modi fications done to the SHAP test protocol in order to be use by the hooks 42 Table 6.2 : Tasks that were unable to be completed by all the test subjects 43 Table 6. 3 : Tasks where the fingertips were used by all of the subjects, by some of them and by no one 44 Table 6.4 : Average, minimum and maximum index of fun ctionality for the Hosmer hook and for the hook with the fingertips each prehensile pattern 55 Table 6.5 : Average index of functionality of the tasks where the fingertips were us ed by everyone 55 Table 6.6 : Average index of functionality of the tasks where the fingertips were not used by anyone 56 Table 6.7 : Average index of functionality of the tasks where some of the subjects used the fingertips 57 Table C.1: Lightweight abstract objects time for subject #1 82 Table C.2: Heavyweight abstract objects time for subject #1 8 2

PAGE 6

v Table C.3 : Activities of daily living time data for subject #1 83 Table C.4 : Lightweigh t abstract objects time for subject #2 84 Table C.5 : Heavyweight abstract objects time for subject #2 84 Table C.6 : Activities o f daily living time for subject #2 85 Table C. 7 : Lightweight abstract objects time for subject #3 86 Table C. 8 : Heavyweight abst ract objects time for subject #3 86 Table C.9 : Activities of daily living time data for subject #3 87 Table C.10 : Lightweight ab stract objects time for subject #4 88 Table C.11 : Heavyweight abstract objects time for subject #4 88 Table C.12 : Activities of daily living time data for subject #4 89 Table C.13 : Lightweight abstract objects time for subject #5 90 Table C.14 : Heavyweight abstract objects time for subject #5 90 Table C. 15 : Activities of daily living time data for subject #5 91 Table C.16 : Lightwei ght abstract objects time for subject #6 92 Table C .17 : Heavyweight abstract objects time for subject #6 92 Table C. 18 : Activiti es of daily living time data for subject #6 93 Table C .19 : Lightweight abstract objects time for subject #7 94 Table C .20 : Heavyw eight abstract objects time for subject #7 94 Table C. 21 : Activities of daily living time data for subject #7 95 Table C .22 : Lig htweight abstract objects time for subject #8 96 Table C.23 : Heavyweight abstract objects time for subject #8 96 Table C .24 : Act ivities of daily living time data for subject #8 97 Table C.25 : Lightweight abstract objects time for subject #9 98

PAGE 7

vi Table C.26 : Heavyweight abstract objects time for subject #9 98 Table C.27 : Activities of daily living time data for subject #9 99 Table C.28 : Lightweight abstract objects time for subject #10 100 Table C.29 : Heavyweight abstract objects time for subject #10 100 Table C.30 : Activity of daily living time data for subject #10 101

PAGE 8

vii List of Figures Figure 2. 1 : Otto Bock hooks [9] 6 Figure 2.2 : Prehensors [10] 6 Figure 2.3: Female passive hand [8] 7 Figure 3.1: CAD models of the (a) f our bar and (b) stiff hinged mechanism 13 Figure 3.2: Model of the crossed four bar 14 Fig ure 3.3: Stability of four bars when a=0.5 19 Figure 3.4: Stability region of four bar mechanism 20 Figure 3.5: Stability re gion of the stiff hinged mechanism when |F/K |=1 21 Figure 3.6: Stability region of the stiff hinged mechanism when |F/K|>1 21 Fi gure 3.7: Stability region of the stiff hinged mechanism when |F/ K |<1 22 Figure 4.1: Fingertips 23 Figure 4. 2 : (a) Lateral and (b) top view CAD model of 555 Hosmer Hook with the fingertips 25 Figure 4.3 : (a) Isometric and (b) side view of hook with fingertips attached 25 Figure 4. 4 : Gripper paddle s used in Wheelchair Mounted Robotic Arm 26 Figure 4.5 : Isometric CAD model of the gripper paddles with the fingertips attached 27 Figure 5.1: Different types of human grasps 30 Figure 5.2: SHAP objects 32

PAGE 9

viii Figure 5.3: Abstract SHAP objects 34 Figure 5.4: Activities of daily living objects 35 Figure 5.5: 555 Hosmer hook 37 Figure 5.6: New design of the ho ok with the fingertips at tached 37 Figure 5.7: CAD model of the gripper paddles used in the Wheelchair Mounted Robotic Arm 38 Figure 5.8: Design of the gripper paddle with the fingertips attached 39 Figure 6.1: Mean normative task times for the non dominant hand 41 Figure 6.2 : Mean times of hooks performing the heavyweight spherical object task for each subject 45 Figure 6.3 : Mean times of hooks performing the lightweight spherical object task for each subject 46 Fi gure 6.4 : Mean times of hooks performing the light weight power object task for each subject 47 Figure 6.5 : Mean times of hooks performing the heavyweight power object task for each subject 48 Figure 6.6 : Mean times of hooks removing the jar lid for each subject 49 Figure 6.7 : Mean times of the hooks pouring water from a jug for each subject 50 Figure 6.8 : Mean times of the hooks pouring water from a carton for each subject 51 Figure 6.9 : Mean times of the hooks moving a jar full of water for each subject 52 Figure 6.10 : Mean times of the hooks moving an empty tin for each subject 53 Figure 6.11 : Mean times of the ho oks rotating a screw for each subject 54 Figure A.1 : Instanta neous stability of crossed four bar mechanism for a=0 66 Figure A.2 : Instanta neous stability of crossed four bar mechanism for a=0.1 66

PAGE 10

ix Fi gure A.3 : Instantaneous stability of crossed four bar mechanism for a=0.2 67 Figure A.4 : Instanta neous stability of crossed four bar mechanism for a=0.3 67 Figure A. 5 : Instanta neous stability of crossed four bar mechanism for a=0.4 68 Figure A.6 : Instanta neous stability of crossed four bar mechanism for a=0.5 68 Figure A.7 : Instanta neous stability of crossed four bar mechanism for a=0.6 69 Figure A.8 : Instanta neous stability of crossed four bar mechanism for a=0.7 69 Figure A.9 : Instantaneous stability of cro ssed four bar mechanism for a=0.8 70 Figure A.10 : Instanta neous stability of crossed four bar mechanism for a=0.9 70 Figure A.11 : Instanta neous stability of crossed four bar mechanism for a=1.0 71 Figure D.1 : Mean values of the spherical prehensile pattern for each subject 102 Figure D.2 : Mean values of the tripod prehensile pattern for each subject 103 Figure D.3 : Mean values of the power prehensile pattern for each subject 104 Figure D.4 : Mean values of the lateral prehensile pattern for each subject 105 Figure D.5 : Mean values of the tip prehensile pattern for each subje ct 106 Figure D.6 : Mean values of the extension prehensile pattern for each subject 107

PAGE 11

x Linkage based Prosthetic Fingertip s : Analysis and Testing Issa A Ramirez ABSTRACT This thesis consists of the research on linkage based fingertips for prosthetic hands The se fingertips consists of small polycentric mechanisms attach ed to what would be the pulp in normal anatomical fingers. The se mechanisms allow the prosthetic hand to conform to the shape of objects during grasp. The goal of these prosthetic fingertips is to maximize the functionality of the hand while minimizing the number of inputs that the user has to control. The stability of the fingertip mecha nisms is analyzed using the principle of virtual work. From this analysis we are able to show that the fingertip mechanism is stable for a large range of rotation of the link and for a large range of directions on which the force is applied, and that the m echanism is indifferent to the magnitude of the force applied to it (assuming that the force does not damage/deform the mechanism). T o assess if the four bar mechanisms (fingertips) improve the grasping capabilities in robotics and prosthetics tests were performed on prosthetic hands and robot grippers with and without the fingertips. Comparisons were made using t he Southampton Hand Assessm ent Procedure (SHAP) protocol, which test s the differences and measure s the functionality of particular types of grasp such as power, spherical, lateral, tripod, tip and extension

PAGE 12

xi In the human testing, t he overall Index of F unctionality (IOF) of the Hosmer hook is 66.65 and 66.21 for the hook with the fingertip s. The hook with the fingertips had a better IOF in the sph erical and power prehensile pattern. When the IOF is calculated for the tasks that the fingertips were used, in 10 of 11 of the tasks, the IOF is higher than using the Hosmer hook. In the robotic gripper testing, the Index of F unctionality was not be calcu lated because the time to perform the tasks depended more on the robotic control system than on the physical characteristics of the gripper.

PAGE 13

1 Chapter 1 Introduction 1.1 Motivation There is a need for prosthetic hands that are simple for the amputee to use and that ease the perform ance of precise manipulation tasks. There has been impressive work done on robot hands, making them humanlike and able to replicate the control of each joint. On the other hand, this work has not been yet put in service to a great extent of the amputee community because the controls required are more complex than those that can be easily supplied by an amputee. The technologies, whi ch are called terminal devices, that are actually used by amputees are either functional or co smetic. The most function al (and least cosmetically appealing ) ar e the hooks The most cosmetically appealing (and least functional) are the cosmetic passive hands (which are basically gloves). The goal of this research is to offer improved f unctionality i n prosthetic terminal devices by attaching mechanisms to a prosthetic hook which will improve its ability to conform to the shape of objects while not impairing its ability to achieve a stable grasp. This research seeks to improve the design of current hoo ks, which may continue to be the

PAGE 14

2 practical choice of amputees for some time. Additionally, there may be some benefit to the work in the design of simple robotic grippers. 1.2 Scope This research de scribes background information o n prosthetic terminal devi ces and describes one mechanism, a crossed four bar, which may improve their functionality. A stability analysis is provided to show how the crossed four bar achieves a stable grasp. The mechanism is implemented in a hook and a robotic gripper. The functi onality of hooks with and without the mechanism is compared using the Southampton Hand Assessment Procedure (SHAP). 1.3 Thesis Overview Information about prosthetic background, including the upper limb prosthetics user priorities is presented in Chapter T wo. Chapter Three consists of the analysis of the crossed four bars and a comparison to a stiff hinged mechanism using the method of virtual work The design of a prosthetic hook and the paddles of the gripper is presented in Chapter Four. Chapter Five pro vides an overview of the method, the SHAP test, used to evaluate the benefits of the fingertips in the hook and in the gripper paddles. Chapter Six presents the results of the paddles with the fingertips compared to the paddles of the Wheelchair Mounted Ro botic Arm, and the hook with the fingertips compared to a commercially available hook. The conclusions and recommendations for future work for this research are presented in Chapter Seven.

PAGE 15

3 Chapter 2 Background This chapter consists of the background of prosthetic hands. It de scribes the causes associated with upper limb amputation and the different types of terminal devices that are currently available. To be able to improve prostheses, the user priorities and the tendencies in design criteria of rec ent research are also presented. 2.1 Epidemiology Statistics of upper limb amputations in the United States from 1 988 to 1996 have shown that 78.1 % of the upper limb amputatio ns are associated to trauma, 3.1 % are congenital, 1.5% are related to cancer an d, 17.3 % are related to dy svascular disease [1]. Although 85,000 90,000 people in the United States have lost an upper lim b, only about 34,000 (40% ) use prosthetic arms or hands [2]. 2.2 Prosthesis A p rosthesis is an artificial substitute or replacement of a part of the body such as a tooth, eye, a facial bone, the palate, a hip, a knee or another joint, the leg, an arm, etc. A prosthesis is designed for functional or cosmetic reasons or both [3]

PAGE 16

4 Functional upper limb prostheses generally can be divided into two categories: body powered prostheses and myoelectric prostheses [4]. Body powered prosthese s are powered and controlled by gross limb movements. These movements, usually of the shoulder, upper arm, or chest are captured by a harness system which is attached to a cable that is connected to a terminal devi ce. Some advantages of the body powered prosthese s are that they are highly durable, and are usually of moderate cost and weight. Some d isadvantages are that the prosthetic u sers feel uncomfortable a nd must use a restrictive control harness. Although new materials aid in reducing discomfort [5] the harness must be tight in order to capture the movement of the shoulder and support the prosthesis. The tight harness can also restrict range of motion and the functional envelope (the area in space where the patient can control his or her prosthesis). Others dislike the look of the hook and control cables and request a prosthesis that is more "lifelike" [5] Myoelectric control uses the electrical signals g enerated by muscle contraction as the control input for a prosthesis controller. They function by transmitting electrical activity that the surface electrodes on the residual limb muscles detect to the electric motor [5] Myoelectric prostheses may give mo re proximal function and increased cosmetics, but they can be heavy and expensive. They have less sensory feedback and require more maintenance. 2.3 Terminal d evices The terminal device is the end effector or prehensor that is situated on the end of the a rm prosthesis [6]. Terminal dev ices are divided in two categories : passive and active terminal devices. The main advantage of a passive terminal device is its cosmetic appearance. With newer advances in materials and design, a device that is virtually

PAGE 17

5 indi stinguishable from the native hand can be manufactured. However, passive terminal devices usually are less functional and more expensive than active terminal devices [4] Active terminal devices usually are more functional than cosmetic; however, in the n ear future, active devices that are equall y cosmetic and functional will be available. Active devices can be broken down into two main categories: 1) myoelectric based devices hooks and 2) prosthetic ha nds with cable A prosthetic hand usually is bulkier and heavier than a hook, but it is more cosmetically pleasing. A prosthetic hand can be powered with a cable or myoelectricity. With the myoelectric device, the patient can initiate palmar tip grasp (to hold smaller objects like pencils) by contracting res idual forearm flexors and can release by contracting residual extensors [4] 2.4 Commercially available terminal devices Currently, there exist three types of terminal devices: hook, prehensors and hands [7]. Each type of terminal device can be either elec tric or body powered. T erminal devices have either voluntary closing or voluntary opening mechanisms. In the voluntary closing mechanism the device is open at rest and activation is required to grip an object. In the voluntary opening mechanism the device is closed at rest and activation is required to open the terminal device Hooks are made of stainless steel or aluminum. Examples of some prosthetic hooks are shown in Figure 2.1. Typical advantages of the hooks are f unctionality, efficiency of use, abilit y to grasp small objects, durability, lower maintenance and repair costs, light weight (compared to hands), better ability to see what the user is trying to hold, and the user does

PAGE 18

6 not have to be as careful around heat because they are made of metal [7] They may be either canted or lyre shaped [8] Canted fingers permit better visualization of the object to grasp. Prehensors (Figure 2.2 ) are not as cosmetically pleasing as prosthetic hands, but they offer many of the same advantages over hands as h ooks do. They are much more functional than passive hands and, like hooks, offer bett er visual feedback to the user. Some a dvantages of the prehensors compared to the hooks is that they d o not look as threatening, they are not likely to scratch objects and are not likely to acciden tally get caught on things. T hey are not as good for picking up and working with small items, and they are usually bulkier at the end which can make it difficult to see the objects being grasped [7] Figure 2. 2 : Prehensors [10] Figure 2. 1 : Otto Bock hooks [9]

PAGE 19

7 Prosthetic hands (Figure 2.3 ) are generally less functional than hooks and prehensors. There are three types of artificial hand mechanisms: passive, voluntary closing and voluntary opening. Passive hands are used for cosme tic reasons, they have the appearance of an anatomical hand; but they need to be opened and closed using the sound hand. Figure 2. 3 : Female passive hand [ 8] The focus of this thesis is in body powered terminal devices. A summary of the commercially available body powered terminal devices is shown in Table 2 1. Table 2. 1 : Functionality and weight of body powere d terminal device s commercially available Manufacturer Name Type Functionality Weight Hosmer [8] Hooks Hook cable activated 85 397 gm APRL Hook cable activated 243 gm Sierra 2 load Hook cable activated 354 gm Soft voluntary closing hand Hand cable activated, vo luntary closing of the thumb and first two fingers 339 351 gm Soft voluntary opening hand Hand cable activated, voluntary opening of the thumb and first two fingers 288 308 gm Dorrance 300 mechanical hand Hand cable activated, voluntary opening o f the thumb and first two fingers 298 gm

PAGE 20

8 Table 2.1: Continued Dorrance 400 mechanical hand H and cable activated, voluntary opening of the thumb and first two fingers 397 gm Becker Lock Grip Hand Hand cable activated, all five fingers o perate 382 467 gm Becker Imperial Hand Hand cable activated, all five fingers operate 393 gm APRL Voluntary Closing Hand Hand cable activated, moveable thumb and two fingers 354 gm Sierra Voluntary Opening Hand Hand cable activated, two thumb po sitions 354 gm Otto Bock [9] Body powered hooks Hook cable activated --------System Hands passive Hand opened by the sound hand and close automatically. 185 290 gm System Hands voluntary opening Hand cable activated 215 340 gm System Hands voluntary closing Hand cable activated 340 380 gm TRS [10] Adult Grip Prehensors Prehensor cable activated 278 451 gm Lite Touch Adult Hand cable activated 284 gm A survey of prosthetic limb users was performed at the O xford Limb Fitting Centre, Headington, Oxford. Many subjects were users of more than one type of prosthesis, includ ing purely cosmetic hands, prosthetic hooks, and devices that have a cosmetic appearance but a limited functional range (such as myoelectric hands) [11] The principal

PAGE 21

9 types of prosthesis that upper limb amputees use are cosmetics (21%), cosmet ics like devices (20%) and prosthetic hooks (13%) only. The cosmetic like devices are those with some designed functional range but an anthropomorphic ap pearance. 2.5 Users priorities A survey of 1,020 body powered prosthetic users was done at The Institute for Rehabilitation and Research (TIRR), Houston in 1992. They evaluate d past use of prostheses, current trends in technology and prosthetic preferences of these individuals to help define future prosthetic research. The results of the trans radial preferences for prosthetic improvements are listed in Table 2. 2 [12] one being the highest priority Table 2. 2 : Trans radial bod y pow ered items for improvement [12] Item name Rank Wrist rotated the terminal device 1 Could do coordinated motions of 2 joints at the same time 2 Wrist moved terminal device from side to side 3 Wrist moved terminal device up and down 4 Required less attention to perform certain functions 5 Could hold small objects better 6 Could hold large objects better 7 Could use it in vigorous activities 8 Weighted less 9 Looked more like a human hand 10

PAGE 22

10 2.6 Design of terminal devices criteria New requirem ents for the design of prosth etic hands in recent research are that they must be l ightweight, compact, reversible ( simplicity of manufacture of both left and right hands), quiet (in the case of myoelectric hands), cos metic (t he device must be attractive), functional, efficient, and price (the device must be produced at a cost that will allow the hand to be priced competitively) [ 13,14,15]. Each hand has its own design criteria and requirements. Design criteria of some terminal devices in research are shown in T able 2. 3. Table 2 3 : Design criteria of some terminal devices in research Design Criteria Prosthetic hands Type Weight Functionality Cosmetics The Southampton hand [13] Myoelectric Yes, n/a Able to grasp everyday obj ects 5 fingers The IOWA hand [18] Myoelectric Yes, n/a Yes, n/a 5 fingers, uses a cosmetic glove The RTR II [19] Myoelectric Yes, n/a Adaptive grasp 3 fingers The Spring hand [14] Myoelectric Yes, n/a Yes, n/a 3 fingers The VA endoskeletal hand [15] Bo dy powered 203 gm Able to grasp everyday objects 5 fingers 2.7 Assistive robot manipulators Upper limb a ssistive devi ces are those devices that help people with limb limitations to perfom activities of daily living. A variety of devices are commercially available. Most of t he devices that currently exist are for a specific task (button hooks for dressing, utensils with adjustable rings, etc.).

PAGE 23

11 There currently exists a number of assistive robot manipulators. The ARM or MANUS, the Handy I, the RAID projec t, Raptor, DeVAR, Robotic Assistive Appliance and the Helping hand are some of these manipulators [16] Some of the end effectors of these manipulators have revealed that some aspects of the gripper have room for improvement. The difficulties experienced by the Assistive Robotic Manipulator (ARM), also known as MANUS are typical. It had pr esent ed difficulty handling large objec ts that weight more than 2 kg. When the center of mass could not be enclosed by the gripper or an external load was applied to it, the objects tend to rotate or tilt. W hen a large object is positioned near the base of the gripper, the two proximal phalanges drive the object out of the g ripper, when the gripper closes [17]. It is possible that the four bar mechanism developed for the prosthetic hand may aid robotic grasp. 2.8 Underactuation A mechanism is said to be underactuated when it has fewer actuators than degrees of freedom. When applied to mechanical hands, underactuated mechanisms can be used to obtain an adaptive grasp that resembles human grasping more easily than a hand with completely independent degrees of freedom could achieve [14] Some underactuated mechanism based prosthetic hand prototypes are the SPRING hand [14] and the Iowa hand [18] which is base d on a c able tra nsmission, the RTR II [19] the Lin ATG prosthetic hand [20] and th e Whole Arm Manipulator (WAM) [21] which connects to the

PAGE 24

12 finger segments via a solid mechanical linkages. A survey of the most well known underactuated robotic hands is shown in [22] Th e crossed four bar mechanism explained in the next chapter is unde ractuated and part of its simplicity comes from the fact that it performs it stabilizing function without the need for external actuation.

PAGE 25

13 Chapter 3 Kinematic Analysis A brief descri ption of equations that can be used to calculate the orientation of the links in the four bar is presented in this chapter Then the analysis of the crossed four ba r and the stiff hinged (Figure 3.1 ) mechanism were developed to find the stable region of ea ch mechanism, which is accomplished by the principle of virtual work. a b Figure 3 1 : CAD models of the (a) Four bar and (b) stiff hinged mechanism 3.1 Analysis of four bar mechanism The lengths of the links are L 1 L 2, L 3, and L 4 (Figure 3.2), and the angles that links 2, 3 2 3 4 For a particular mechanism with fixed geometry, given any of the angles, the other two can be found by solving the equations:

PAGE 26

14 ( 3. 1) ( 3. 2) For any given position of the four bar, the method of virtual work can be used to determine the conditions in which the four bar will be in equilibrium. Figure 3 2 : Model of the crossed four bar 3.2 Principle of virtual work The principle of virtual work states that the net virtual work of all active forces is zero if and only if an ideal mechanical sy stem is in equilibrium [2]. The total virtual work of the system can be written as ( 3. 3)

PAGE 27

15 Where : is the virtual work F is the applied force vector M is the applied moment vector V is the potential energy is the virtual displacement is the virtual angular displacement and q is the generalized coordinate. The analysis of the mechanism is developed using the method of virtual work [10]. The position vector of the force applied to th e crossed four bar linkage (Figure 3.2 ) is given by : ( 3. 4) A fictitious moment, M out is applied as a measure of how far the four bar is away from equilibrium given a force acting in the direction, applied at a fraction, a of the length of link 3 L 3 3.3 C rossed four bar mechanism The virtual work due to the force is found by applying the dot product of the force vector and the virtual displacement. Given the force, F the virtual work due to the applied force is found to be ( 3. 5)

PAGE 28

16 Where: ( 3. 6) is the four bar kinematic coefficient [10], is the angle of the force a is the distance at which the force is applied, as shown in Figure 3 2. The virtual work due to the moment is found by applying the dot product of the moment and the virtual angle displacement. ( 3. 7) The net virtual work is the sum of the virtual work due to the force and due to the moment is found to be ( 3. 8) From the principle of virtual work (if equilibrium, the virtual work is equal to zero) the moment on link 3 of the cro ssed four bar mechanism required for equilibrium is found to be ( 3. 9) When M ou t =0, the crossed four bar is at equilibrium which occurs when the force angle, satisfies ( 3. 10) which does not depend on the magnitude of the force, F

PAGE 29

17 3.4 S tiff hinged mechanism Using the principle of virtual work, the work due to the torsional spring is ( 3. 11) Using the principle o f virtual work for the stiff hinged mechanism, as in the crossed four bar mechanism, the moment is found to be ( 3. 12) where k is the spring stiffness. When M out =0 in the stiff hinged mechanism ( 3. 13) This is a definite contrast to the four bar mechanism in that equilibrium in the stiff hinged mechanism depends on the ratios of the spring stiffness and the applied force. 3.5 S tability analysis Using the moment equation, equation ( 3. 9 ), M out as a function of the angle of link 3, 3 and the angle force, can be found. The equilibrium moment of the crossed bar is found for different locations of the applied force by changing the value of distance a Appendix A shows different plots of the angle of link 3 an d the angle force for different values of the magnitude a and for the geometric parameters stated in Table 3. 1. Table 3. 4 : Geometric p arameters Set L 1 L 2 L 3 L 4 1 1.04 1.125 1 1.125

PAGE 30

18 The curves represent the moment that would be r equired to balance the applied force. Equilibrium curves for the different mechanisms were developed using the angle of the link, the force angle and a moment created by the force. Each equilibrium curve represents a different position on the coupler link of the four bar (distance a ). Figure 3.3 shows the plot of the equilibrium curve when a =0.5, i.e. the center of link 3, in a range from 0 o 360 o Equilibrium without an applied moment is achieved when the moment out of the mechanism is equal to zero. Wh en the mechanism is loaded at 90 o (pulling on the mechanism) in a way that changes the orientation of link 3 from a value of 180 o (horizontal) to 200 o (slightly tipped), this corresponds to moving from point A to point B in Figure 3.3 At this position a m oment, M out of 5 Nmm, assuming an applied force of 1 N, would be required to hold it in equilibrium. Because there is nothing in the mechanism to provide that moment the linkage will tend to continue to rotate in the positive direction and the perturbatio n will grow i.e. 3 will continue to get bigger. We say that this situation is unstable because even a very small perturbation will grow until In the other hand, when the applied force is push ing in the vertical dir =270 o on the center of link 3 and link 3 is perturbed from 180 o to 200 o (from C to D in Figure 3.3 ), the resulting force is 5 Nmm, indicating that this unbalanced moment will cause link 3 to rotate in the negative direction, back to its original position. Because small perturbation tend s to be resisted, we say that this equilibrium position is stable when negative values of moment are above an equilibrium contour line (where M ou t =0) and positive values of moment are below the line. An exam ination of the plots in Appendix A shows that when the slope of equilibrium contours line at a point is negative (up to the left, down to the

PAGE 31

19 right), the equilibrium point is stable. On the other hand, when the slope of an equilibrium contour line is positive (up to the right, down to th e left) the equilibrium point is unstable. Figure 3. 3 : Stability of four bars when a=0.5 The stable region for the crossed four bar mechanism is the shaded region in Figure 3.4 The advantage of this mechanism is that the magnitude of the force does not change the stability of the mechanism and it is stable over a wide range of motion and for a wide range of applied force directions and locations. (degrees) (degrees)

PAGE 32

20 (degrees) Figure 3 4 : Stability region of four bar mechanism In the stiff hinged model, the magnitude of the force plays a major role in the stability of the mechanism, the mechanism presents 3 different types of stability depending on the ratio of the force to the spring stif fness. Figure 3.5 shows the stability when the ratio of the force to the spring stiffness is equal to one. (degrees)

PAGE 33

21 Figure 3. 5 : Stability region of the stiff hing ed mechanism when |F/K |=1 Figure 3. 6 : Stability region of the stiff hinged mechanism when |F/K|>1 |F/K|>1 |F/K|=1

PAGE 34

22 Figure 3. 7 : Stability region of the st iff hinged mechanism when |F/ K |<1 Figure 3.6 shows the stability when the magnitude of force is bigger than the spring stiffness (F/K >1). When the magnitude of the force is less than the spring stiffness magnitude (F/K <1), Figure 3.7, the stable region decreases and tends to be dominated by the torsional spring. The disadvantage of the stiff hinged mechanism is that the degree to which it conforms to the shape of an object depends on the grasping force. |F/K|<1

PAGE 35

23 Chapter 4 Design In order to measure the benefits of the finge rtips, the fingertips are compared to existing terminal devices. For this, a prosthetic hook and the paddles of a gripp er were designed. This chapter consists of the design procedure for the hook and the paddles of a robotic gripper. 4.1 Fingertips The bars of the crosse d four bar are pinned connected, as shown in Figure 4.1 The bar length s of the fingertips are L 1 =L 3 20 .01mm and, L 2 =L 4 = 21.6 mm. The fingertips have a rubber gripping surface that is 2 mm wide. This makes an overall height of 12.50 mm. Figure 4. 1 : Fingertips

PAGE 36

24 4.2 Hooks One of the commercially available hook s is the Hosmer Dorrance 555 hook Some characte ristics of this hook are [8 ]: 1. I t i s made of a luminum 2. It h as lyre shaped fingers which allows the user to grasp small objects with the tip of the tine and ease in picking up cylindrical objects 3. The fingers are nitrile rubber lin e 4. I t i s 13.3cm (5 lo ng 5. It w eights 4.5oz (121g m ) 6. I t is generally a light duty device To be able to measure if the fingertips possess some kind of benefit, the fingertips will b e attached to a hook. Before the fingertips are placed on the hook, a CAD model was de veloped to de termine the best location f or the fingertips A CAD model of the 555 Hosmer hook is shown in Figure 4. 2 One of the problems of attaching the fingertips to the existing hook is that there is not enough space to place them. An examp le of this is shown in F igure 4. 2 the fingert ips are intersecting e ach other and there is no space to place them because the hook is mostly curved.

PAGE 37

25 Figure 4. 2 : (a) Lateral and (b) top view CAD model of 555 Hosmer h ook with the fingertips Because of lack of space in t he normal hooks, another prosthetic hook was designed to meet with the specifications of the fingertip. Figure 4 .3 shows a cad model of the hook with the fingertips attached. Some changes in this des ign is that the designed hook is 154.7 mm of length and is 4 7.46 mm wide. The inside of the hook is wider (47.46 mm vs. 18 mm) than the Hosmer hook, therefore, there is more space when the fingertips are attached. By implementing this change, the distal part of the hook is able to have the fingertips together when the hook is closed and other fingertips are 10 mm apart. a b Figure 4. 3 : (a) Isometric and (b) side view of hook with fingertips attached a b

PAGE 38

26 4.2.1 Manufacture There were diffic ulties in the manufacture of the prosthetic hook, for this reason another prosthetic hook was developed. This design is 16.3 cm long and has a maximum inside width of 34 mm to attach the fingertips. With that inside width, there is only room to attached 2 fingertips in each side of the hook. 4.3 Robot gripper paddles A robotic gripper, shown in F igure 4. 3, was designed for Wheelchair Mounted Robotic Arm developed at USF. It has shown to be able to grasp different door handles ( knob and lever handles), she ets of paper, a ball, a cup and a rectangular object [ 31 ]. Figure 4. 4 : Gripper paddles used in Wheelchair Mounted Robotic Arm The Wheelchair Mounted Robotic Arm does not have sufficient space to incorporate the fingertips to it. For this reason, an alt ernative robotic paddle for the gripper i ncorporating the fingertips was designed.

PAGE 39

27 Figure 4. 5 : Isometric CAD model of the gripper paddles with the f ingertips attached The paddles are made of aluminum and are 114.30 mm (4 in) height and 50.8 mm (2 in) w ide. This gripper consists of six fingertips (3 on one side and 3 on the other side of the gripper) in parallel, as shown in F igure 4 .4, and has a similar size of the paddles used for the Wheelchair Mounted Robotic Arm. The top of each fingertip is adjacen t to another fingertip on the othe r paddle. This configuration should make the gripper be able to pick up small objects better

PAGE 40

28 Chapter 5 The SHAP T est For thi s, the SHAP test, a standardized hand assessment, will assist in the evaluation of the hooks. First, the different types of human prehensile patterns used in the hand assessment procedure are explained. Each type of prehensile pattern will be used to provi de the index of functionality of the different hooks, comparing them with the anatomical hand. 5.1 P rehensile patterns The major function of the hand that a prosthesis tries to replicate is grip [4 ]. Two different functional types of grasps are differentia ted: power grasp and precision grasp [ 25 ]. Power grasp includes cylindrical ( or power ) spherical ( or flexion ) hook (or extension grip) and lateral prehension when the thumb is adducted. Precision grip includes palmar prehension (or tripod grip), tip and lateral prehen sion when the thumb is abducted [26 ]. The hand forms of each one of these types of grasp is s h own in Figure 5. 1. The general ch aracteristics of each prehensile pattern are as follows: 1) Spherical grip: all fingers and the thumb are flexed, rot ated and abducted to surround and support the object. Also categorize d as the five fingered grip [28 ]

PAGE 41

29 2) Tripod pinch: the pulp of the thumb is opposed to the pulp of the index and middle fingers. 3) Power grip: all fingers are flexed around the object. The p alm of the hand is used for object opposition rather than active force generation by the thumb. Also categorized as diagonal and transvolar, fir st, cylindrical or hook grips [25 28 ]. 4) Lateral pinch: A small, flat object is held between the lateral (radial) aspect of the middle or distal phalanx of the index finger and the pulp of the thumb [ 27 ]. 5) Tip pinch: The pulps or the volar aspects of the fingers including the pulps and the pulp of the thumb surround the object and support it against its center. Only on e or two of the radial fingers and the thumb participate and the contact areas are limited to the tips [ 27 ]. 6) Extension grip: all fingers are extended and adducted with the thumb in extension and opposition [ 28 ].

PAGE 42

30 Figure 5.1: Different types of human gra sps [ 29 ] The difference between the two broader categories of grasp is that the hand position for the power grasp is static (fixed position) and for precision grip the hand position is dynamic (for object manipulation) In Figure 5.1, the power ( b ), exten sion ( e ), lateral ( a ) and spherical ( f ) prehensile patterns correspond to a power grasp because the hand position is fixed. The tip and tripod grasps, Figure 5.1 d and c respectively, are used for manipulation, for this reason they are part of the precisio n grasp. 5.2 Southampton h and a ssessment p rocedure The purpose of the Southampton Hand Assessment Procedure (SHAP) is to determine the effectiveness of a terminal device and controller by focusing the evaluation in the unilateral performance of the user. The SHAP consists of twenty six (26) timed tasks;

PAGE 43

31 twelve (12) are abstract tasks and fourteen (14) consist of activity of daily living (ADL) (Figure 5.2) This assessment tool is designed to be a standardized and objective method of evaluating pathologic or prosthetic hand function The SHAP test has undergone validation and reliability studies [29 ] which support its use as an objective assessment tool The abstract objects tasks evaluate prehension without the complication of the tools or equipment used during that often cause intermediate grip patterns or adverse evaluation effects. Because the shape and form of an ADL task is likely to be known to the subject, a psychologic prejudice may exist as to the ability to perform the task. The abstrac t objects remove such an effect to a limited extent. The abstract objects are produced in two sets for use in the SHAP procedure. The first is made of noncompliant dense materials (heavyweight abstracts) and the second is from marginally compliant, low den sity materials (lightweight abstracts), to produce a difference in both weight and yield [29] Colin Light [28 ] divided the fourteen (14) activities of daily living in the six prehensil e patterns as shown in Table 5.1 A spherical grip is required for 10% of the tasks, a tripod grip for 10%, a power grip for 25%, a lateral grip for 20%, a tip grip for 20%, and an extension grip for 10%. This configuration ensures that a full range of natural grips will been evaluated.

PAGE 44

32 Figure 5.2: SHAP o bjects For these t asks, the opposite hand acts only as a stabilizer, thereby ensuring the functional assessment for the impaired hand. The self timed nature of the SHAP eliminates the need for subjective opinion by the asse ssor. Table 5. 1 : SHAP Task 1 Pick up coins Tip 2 Undo buttons Tip/Tripod 3 Simulate food cutting Tripod/Power 4 Simulate page turning Extension 5 Remove jar lid Spherical 6 P our water from jug Lateral 7 Pour water from carton Spherical 8 Move a jar full of water Power 9 Move an empty tin Power 10 Move a tray Lateral/Extension 11 Turn a key Tip/Lateral 12 Open/close a zipper Tip/Lateral 13 Rotate a screw Power 14 Turn a door handle Power

PAGE 45

33 5.3 SHAP Test Personal and prosthetic history was collected from each person. The information i ncluded : age at time of the assessment, ethni c category and gender The subjects were ask ed to perform the SHAP using their hand, a standa rd hook and the hook with the fingertips attached. The detailed protocol is stated on Appendix B The hand results will be used to obtain the normative data for the test. A boundary condition of 8 times that of the norm is imposed in each task of the SHAP test to prevent a subject from taking too long [29]. This definite, if arbitrary, limit on the time allowed to perform a task, provides a numerical value that can be used when subjects are unable to perform a task. 5.3.1 Abstract o bjects Subjects pick ed up each abstract object from the SHAP board (Figure 5.3) move d it through an obstacle and place d it on the SHAP board. For these tasks, the opposing hand acts only as a stabilizer. Time for each task ends when the object is placed on the SHAP board. The tim e was taken by the participant and recorded by the assessor. Each task was performed three times. The abstract objects are: 1. lightweight spherical 2. lightweight tripod 3. lightweight power 4. lightweight lateral 5. lightweight tip 6. lightweight extension 7. heavyweight sph erical

PAGE 46

34 8. heavyweight tripod 9. heavyweight power 10. heavyweight lateral 11. heavyweight tip 12. heavyweight extension Figure 5.3: Abstract SHAP objects 5.3.2 Activities of daily living Subjects perform ed different activities of daily living (Figure 5.4). The time was be taken by the participant and record ed by the assessor. Each task was be perfor med three times. The tasks considered were to : 1. pick up coins 2. undo buttons 3. simulate food cutting 4. simulate page turning 5. remove a jar lid 6. pour water from jug

PAGE 47

35 7. pour water from carton 8. move a jar full of water 9. move an empty tin 10. rotate a screw 11. turn a door handle 12. open and close a zipper 13. turn a key 14. move a tray Figure 5.4 : Activities of daily living objects 5.4 Index of functionality The score given by the SHAP test is a functional s core; 100% indicating normal hand function, made up of six sub scores for each of the different hand grips: lateral, power, tripod, tip, extension and spherical. The Index of Functionality (IOF) which provides an original metric capable of distinguish ing between levels of function, may be obtained for each prehensile pattern, which themselves are comprised of multiple tasks.

PAGE 48

36 Colin Light [28] creates the Index of Functionality using the Euclidean squared distance, a measure between samples in an i dimensi onal problem, where i patterns) in this case. This distance d (equation 15 ) is determined using the z value (equation 14 ) for each prehensile pattern. ( 5. 1 ) Where: x i is the is the mean time for pattern i in the normative sample data taken from healthy subjects s i is the standard deviation of times for the pattern i in the normative sample ( 5. 2 ) The index of functionality (dis tance d ) is rescaled to a value of 100 when x i is equal to the corresponding and diminishes to 0 for a subject that reaches the boundary condition for each task (and hence is deemed to have a minimal function) [28]. The rescaled in dex of functionality (equation 16) is found using the slope point equation. ( 5. 3 ) Where the 7 in the equation 5.3 comes from the boundary condition where x i is 8 times the mean time of the normative sample. The index of functional ity of the standard hook and the hook with the fingertips will be generated to provide an overall assessment of their relative function.

PAGE 49

37 5.5 Human s ubjects For this research, subjects between the ages o f 18 to 25 years were selected. Some reasons for this age i nterval is the convenience at university setting, a normative database has been established as the benchmark of normal hand function in that age range, and that it reduces a potentially significant variation within the group. A previous study has sho wn a slight increase in the time to complete the task with age [ 30 ]. The participants are going to do the different tasks of the procedure with their anatomical hand, the standard hook ( F igure 5.5 ), and with the hook with the fingertips (F igu re 5.6 ), for t hree times each. With these times, the index of functionality of both hooks and the anatomical hand will be generated. Figure 5. 5: 555 Hosmer hook. Figure 5 6: New design of the hook with the fingertips attached.

PAGE 50

38 5.6 Robot gripper A robotic gripper, shown in F igure 5. 7 was designed for Wheelchair Mounted Robotic Arm developed at USF. It has shown to be able to grasp different door handles (knop and lever handles), sheets of paper, a ball, a cup and a rectangular object [ 31 ]. Figure 5. 7 : Cad model of the gripper paddles used in the Wheelchair Mounted Robotic Arm An alternative robotic gripper incorporating the fingertips has been designed (Chapter 4 ). This gripper consists of six fingertips (3 on one side and 3 on the other side of t he gripper) as shown in F igure 5. 8 and has a similar size of the paddles used for the Wheelchair Mounted Robotic Arm.

PAGE 51

39 Figure 5. 8: Design of the gripper paddle with the fingertips attached Both grippers wer e compared using the SHAP test, they attempt ed to do th e same tasks as the human subjects. A boundary condition of 8 times that of the norm was imposed in each task of the SHAP test Because the time to complete the tests depend s on the ability of the person t o manipulate the robot gripper, the benefits of the fingertips will be distinguished by comparing which objects each one of the grippers can grasp.

PAGE 52

40 C hapter 6 Results This chapter presents the results of the human and robotic testing. The SHAP test was used to compare both hooks. A group of 10 pers ons (8 males and 2 females) were asked to performed the SHAP test. All the persons were right handed and nine of them are between the ages of 18 to 25. They used a pseudo prosthesis to simulate a prosthetic user. Using this, they were able to use the conve ntional hook (555 Hosmer hook) and the hook with the fingertips. 6.1 Normative data Each subject performed the procedure specified in Appendix B with their left hand to dev eloped the normative data. The normative data for the hand, Figure 6.1, was used to calculate a range of the index of functionality for the hooks. The error bars in the normative data represents the twice the standard deviation in each task. 6.2 Hook testing The SHAP test is a standardized test for the human hand. Some modifications w ere done in the SHAP protocol to be able to perform these tasks using the hooks.

PAGE 53

41 Figure 6.1: Mean normative task times for the non dominant hand 41

PAGE 54

42 T able 6. 1 : Modifications done to the SHAP test protocol in order to be used by the hooks Tasks Modif ications Tip and Extension Objects The form board is not going to be directly in front of the person. The person can move it to the left or right, in such a distance that he or she is able to perform the task. Lateral Objects When trying to pick up this object parallel to the handle, as done with the hands, the object tends to rotate and falls. Because of this problem, the object could be lifted perpendicular to the handle. Simulate page turning To be able to pick up the index card, the person needs to drag it, using the hook, below the mark. In this way, the hook is able to pick it up. Pour water from jug, pour water from carton, move a jar full of water and move an empty tin In these tasks, the form board is not going to be directly in front of the pe rson. The person can move it to the left or right, in such a distance that he or she is able to perform the task. 6.3 Observations There is a time difference in subjects 9 and 10, which are females. This is because men generally tend to move more rapidl y because of the greater amount of contra ctile force as noted in [32 ]. Another parameter is that subject 10 is over the age specified and with an increase in age, there is an increase in timing. The data, Appendix C, was divided in their correspondent pre hensile patterns. Appendix D, shows a plot of the mean values of each person compared to themselves using their hand, the Hosmer hook and the hook with the fingertips.

PAGE 55

43 6.4 Data analysis There were difficulties completing some of the tasks. In the test, th e participants are using their no n dominant hand. Table 6.2 lists tasks that were not able to be completed by all the subjects. Table 6. 2 : Tasks that were unable to be completed by all the test subjects Task Hook (tasks completed/ total tasks) Hook with fingertips (tasks completed/ total tasks) Lightweight Lateral 10/10 9/10 Heavyweight Spherical 8/10 10/10 Heavyweight Power 9/10 10/10 Heavyweight Lateral 10/10 9/10 Undo buttons 9/10 6/10 Simulate food cutting 5/10 2/10 Remove jar lid 7/10 10/10 O pen/close a zipper 10/10 7/10 There is more than one way to grasp an object with the hooks. Not everyone had the same approach grasping the objects. Table 6.3 shows the tasks for which the four bars were used by every subject, the tasks in which some use d the four bars and by not anyone. The tasks where the four bars where not used by anyone, the objects are small. The tasks where the four bars where used by everyone, the objects are large. We are going to be focus on the tasks that the fingertips were us ed by everyone.

PAGE 56

44 Table 6. 3 : Tasks where the fingertips were used by all of the subjects, by some of them and by no one Four bars used by everyone Four bars used by some subjects Four bars not used by anyone Lightweight Spherical Lightweight Tripod Lig htweight Tip Lightweight Power Lightweight Lateral Lightweight Extension Heavyweight Spherical Heavyweight Tripod Simulate page turning Heavyweight Power Heavyweight Lateral Turn a key Remove jar lid Heavyweight Tip Open/close a zipper Pour water from jug Heavyweight Extension Turn a door handle Pour water from carton Pick up coins Move a jar full of water Undo buttons Move an empty tin Simulate food cutting Rotate a screw Move a tray In the heavyweight spherical obj ect task, Figure 6.2, two s ubjects were not able to performed the task. The plot shows that six subjects were able to complete the task in less time taken using the hook with the fingertips. Just one subject had a better timing using the Hosmer hook.

PAGE 57

45 Figure 6. 2 : Mean times of hooks performing the h eavyweight spherical o bject task for each subject Everyone was able to perform the lightweight spherical task, Figure 6.3. Five of the subjects had a significant difference in the time to perform the tasks using the hook wi th the fingertips. Two subjects had better timing using the Hosmer hook.

PAGE 58

46 Figure 6. 3 : Mea n times of hooks performing the l ig htweight spherical o bject task for each subject In the lightweight power task, Figure 6.4, six subjects had a better timing using the hook with the fingertips, two subjects did better with the Hosmer hook and two subjects do not show a significant difference in the time.

PAGE 59

47 Figure 6. 4 : Mean times of hooks performing the lightweight power o bject task for each subject In the heav yweight power task, Figure 6.5, one subject was not able to perform the task using the Hosmer hook, it was too heavy and could not pick it up. All the subjects had a better time using the hook with the fingertips.

PAGE 60

48 Figure 6. 5 : Mean times of hooks perf orming the h eavyweight power o bject task for each subject Seven persons were able to remove the jar lid with the Hos mer hook, as shown in Figure 6.6 Of the seven subjects that completed the task with both hooks, five subjects had a better time using the hook with the fingertips, one subject had a better timing using the hook and one person do not show any significant difference.

PAGE 61

49 Figure 6. 6 : Mean times of hooks removing the jar lid for each subject When the subjects were pouring water from the jug, six subjects had better time using the hook with the fingertips, one had better time using the Hosmer hook, and three subjects do not show any significant difference as shown in Figure 6.7 The Hosmer hook had a standard deviation of 5.27 and the hook with th e fingertips of 3.55.

PAGE 62

50 Figure 6. 7 : Mean times of the hooks pouring water from a jug for each subject In the task of pouring 200 ml of water from a carton, five subjects had a better timing using the hook with the fingertips, three subjects using the Hosm er hook, and two subjects do not show any significant difference as shown in Figure 6.8 For this tasks, subjects 2,3,5,6,7,9, and 10 rotated the carton 90 degrees and then picked it up. In t his way, they did not need to use much force in opening the ho oks.

PAGE 63

51 Figure 6. 8 : Mean times of the hooks pouring water from a carton for each subject S ome problems were encountered lifting the jar full of water with th e conventional hook. When the subjects trie d to lift the jar by the base, the objects tended to des cend (caused by the weight and friction on the fingers). Most of th e subjects grasped the jar in between the lid and the base. This helped them to secure the object before lifting it. As shown in Figure 6.9, seven subjects had better timing using the hook with the fingertips, two subjects had better time using the Hosmer hook and one did not show any time difference.

PAGE 64

52 Figure 6. 9 : Mean times of the hooks moving a jar full of water for each subject In moving the empty tin, Figure 6.10 five subjects had b etter time using the hook with the fingertip s, two subjects had better time using the Hosmer hook, and three did not show any significant difference.

PAGE 65

53 Figure 6. 10 : Mean times of the hooks moving an empty tin for each subject To rotate the screw, with the hook with the fingertips, the subjects dragged down the screwdriver, and picked it up using the last 2 fingertips. Figure 6.11 shows that s even subjects had better time using the hook with the fingertips, and three had better time using the Hosmer hook.

PAGE 66

54 Figure 6. 11 : Mean times of the hooks rotating a screw for each subject 6.4 Index of functionality The overall Index of F unctionality (IOF) of the Hosmer hook is 66.65 (range d from 46.25 to 80.03) and 66.21 (range d from 47.15 to 77.42) for the hook with the fingertips. One caveat in interpreting the IOF is that some of the tasks could not be completed by some subjects. This increases the timing, (because the boundary condition of 8 times the mean normative value is used ), and decreases the IOF. Table 6. 4 shows the Index of F unctionality for each prehensile pattern. In t he spherical and power grasps, the hook with the fingertips had a better average IOF This is also because in at least in one of the tasks of the other prehensile patterns, one or more sub jects could not perform it.

PAGE 67

55 Table 6. 4 : Average, m inimum and m aximum i ndex of f unctionality for the Hosmer hook and for the hook with the fingertips for each prehensile pattern Hosmer Hook with fingertips Prehensile pattern a verage min m ax a verage m in max Spherical 61.43 35.28 87.04 77.64 63.41 86.79 Tripod 54.58 34.70 63.60 46.85 31.18 56.51 Power 65.81 44.69 74.75 70.98 56.76 77.68 Lateral 73.94 52.53 86.92 69.63 40.44 85.15 Tip 67.26 48.08 79.98 56.58 31.66 76.18 Extension 73.42 57.59 88.70 69 .66 47.28 86.03 The mean Index of Functionality of the tasks for which the subjects always used the fingertips, Table 6.5, shows that the hook with the fingertips was able to perform 10 of 11 of the tasks in less time than using the fingertips. An except ion of this was moving the tray. The most significant improvements are in the heavyweight spherical task, heavyweight power task, removing the jar lid and pouring water from a jug. Table 6. 5 : Average index of f unctionality of the tasks where the fingerti ps were used by everyone Task Hosmer Hook with fingertips Lightweight Spherical 73.35 77.60 Lightweight Power 75.18 79.11 Heavyweight Spherical 54.81 74.25 Heavyweight Power 52.63 79.07 Remove jar lid 34.34 73.18 Pour water from jug 77.26 83.22 Pour water from carton 83.23 85.52 Move a jar full of water 76.81 79.24 Move an empty tin 80.92 84.52 Move a tray 72.43 71.32 Rotate a screw 75.79 79.10

PAGE 68

56 The mean Index of Functionality of the tasks w h ere any subject used the fingertips, Table 6.6, the i mportance is the shape and material of the hook. For the page turning, turn a key and to open and close a zipper tasks, the subjects used the inner side and the tips of the hooks. The significant difference in the task of opening and closing the zipper, t he problem in the hook with the fingertips is that it does not have any friction material in the inner side of the finger, and the Hosmer hook has a rubber coating at the tips. Table 6. 6 : Average index of f unctionality of the tasks where the fingertips were not used by anyone Task Hosmer Hook with fingertips Lightweight Tip 73.02 76.94 Lightweight Extension 81.07 82.00 Simulate page turning 70.80 70.91 Turn a key 77.81 79.22 Open/close a zipper 61.22 36.17 Turn a door handle 87.18 86.42 The me an Index of Functionality of the tasks that not all t he subjects used the fingertips, Table 6.7, shows that the hook with the fingertips was able to performed 3 of 9 of the tasks in less time than using the fingertips.

PAGE 69

57 Table 6 .7 : Average index of f unctionality of the tasks where some of the subjects used the fingertips Task Hosmer Hook with fingertips Lightweight Tripod 73.65 77.61 Lightweight Lateral 77.06 76.31 Heavyweight Tripod 72.48 77.15 Heavyweight Lateral 77.84 71.58 Heavyweight Tip 70.51 69.51 Heavyweight Extension 77.19 78.35 Pick up coins 61.00 54.43 Undo buttons 60.00 23.20 Simulate food cutting 12.17 9.44 Some uncontrolled variables in the testing are: 1) Strength and agility of the person: some of them are quick learners an d are able to perform the tasks easily Upper body strength helps them to open the hook faster. 2) Frustration: there is a level of frustration that the person has when is not able to co mplete a task. Some of the activities of daily living are difficult to pe rformed using the hooks. These tasks are picking up coins, t o undo buttons, simulate food cutting, pouring water and opening and closing the zipper 3) Rubber bands: the rubber bands were changed after every two subject to reduce their tendency to degradate o ver time. 6.5 Robotic g ripper t esting The robotic gripper was test using the Manus robotic arm. The movement of the arm was controlled using the Phantom, and the closing and opening of the gripper was controlled using the user interface of the wheel chair mounted arm. This change was done because

PAGE 70

58 there were problems controlling the movement of the arm with the Wheelchair Mounted Robotic Arm (WMRA) system. The paddles of the WM RA gripper have a smooth finish. The gripper could grasped all the items in the s et of lightweight objects. The heavyweight objects are made of steel, and the paddles do not have sufficient friction to be able to grasp the heavyweight object. When using the fingertips, the paddles could perform all the abstract objects tasks with the exception of the heavyweight lateral object. In this task, the object rotates when the gripper was ascending and it was dragged to the end position for the task.

PAGE 71

59 Chapter 7 Conclusions and Recommendations 7.1 Conclusion In the human testing, t he me an Index of F unctionality (IOF) of the Hosmer hook is 66.65 and 66.21 for the hook with the fingertip s. The hook with the fingertips had a better IOF in the spherical and power prehensile pattern. This was because in the other prehensile patterns one or m ore subjects could not perform one or more tasks and the subjects did not use the fingertips in all the tasks. When the IOF is calculated for the tasks that the fingertips were used, in 10 of 11 of the tasks, the IOF is higher than using the Hosmer hook. I n the roboti c gripper testing, the I ndex of F unctionality was not be calculated because the time to perform the tasks depended more on the robotic control system than on the physical characteristics of the gripper. 7.2 Recommendations for future work 1. Anal ysis of the four bar mechanism with other dimensions to increase the stability region.

PAGE 72

60 2. The fingertips had shown that they can conform to different shapes. The functionality of the hook with the fingertips will increase if it had a friction surface at the tips or having small fingertips all over the inner side of the finger. This will help in picking up small objects 3. For the robotic gripper, another testing may be using the theory of contact points. But one observation is that it will be good to incorpora te a friction surface in the WM R A paddles to increase the materials that the gripper can grasp. 4. Detailed statistical analysis of the results.

PAGE 73

61 References [1] Dillingham, T.R. (2002). Limb Amputation and Limb Deficiency: Epidemiology and Recent Tr ends in the United States Southern Medical Journal V ol 95 pp 875 83 [2] Frey, D.D., Car lson, L.E., and Ramaswamy, V., (1995). Voluntary Opening Prehensors with Adjustable Grip force Journal of Prosthetics and Orthotics. Vol. 7. Num. 4 pp 124 13 1. [3] Definition of Prosthesis (2003). MedicineNet.com http://www.medterms.com/script/main/art.asp?articlekey=5076 [4] Martinez, K. Mipro, R C., and Bodeau, V S., Upper Limb Prosthetic s eMedicine. February 13. [5] Kulley, M., (2003). Hand Prosth etics April 28. [6] Muilenburg, A.L., and LeBlanc, M.A. Body Powered Upper Limb Components. Chapter 5. pp. 30. [7] Prosthetic Devices for Upper Extremity Amputees. Amputee Coalition of America in partnership with the U.S. Army Ampu tee Patient Care Prog ram, p p. 46 48 [8] Hooks, Hands, and Wrists. Hosmer Dorrance Corporation http://www.hosmer.com. [9] Otto Bock www. ottobockus.com [10] TRS 2007 catalog http://www.oandp.com/products/trs/.

PAGE 74

62 [11] Kyberd, P.J., Dav ey, J.J., and Morrison, J.D. (1998 ). A Survey of Upper Limb Prosthesis Users in Oxfordshire Journ al of Prosthetics and Orthotics Vol. 10 Num. 4. pp. 85 91. [12] Atkins, D.J., Heard, D.C.Y., and Donovan, W.H., (1996). Epidemiologic Overview of Individuals with Upper Limb Loss and The ir Reported Research Priorities Journal of P rosthetics and Orthotics. Vol. 8. Num. 1. pp. 2 11. [13] Kyberd P.J., Light C., Chappell P.H., Nightingale J.M., Whatley D., and Evans M. (2001). The design of anthropomorphic prosthetic hands: A st udy of the Southampton Hand. Robotica 19. pp. 593 600 [14] Carrozza, M.C., Suppo, C., Sebastiani, F., Massa, B ., Vecchi, F., Lazzarini, R., Cutkoski, M.R., and Dario, P., The Spring Hand: Development of a Self Adaptive Prosthesis for Restoring Natural Grasping. Autonomous Robots. Vol. 16. pp. 125 141. [15] Dosh i, R., Yeh, C., and LeBlanc, M., (1998). The design a nd development of a gloveless endoskeletal prosthetic hand Journal of Rehabilitation Research and Developme nt. Vol. 35. No. 4. pp. 388 395. [16] Kumar, Rahman and Krovi. (1997). Assistive Devices For People With Motor Disabilities Wiley Encyclopaedia of Electrical and Electronics Engineering. [17] Rmer, G., Stuyt, H., Kramer, G., M., and Scheffe J., (2005). Alternative grippers for the Assistive Robotic Manipulator (ARM). Proceedings of the 2005 IEEE 9th International Conference o n Rehabilitation Robotics June 28 July 1. Chicago, IL, USA [18] Yang J., Pitarch E.P., Abdel Malek K., Patrick A., and Lindkvist L. (2004). A multi fingered hand prosthesis Mech anism and Machine Theory. Vol. 39 pp. 555 581. [19] Massa, B., Ro ccella, S., Carrozza, M.C., and Dario, P. (2002). Design and development of an underactuated prosthetic hand Proceedings. ICRA '02. IEEE International Conference on Robotics and Automation Vol.4. pp. 3374 3379. [20 ] Chang, W.T., Tseng, C.H. and Wu, L.I. (2004). Creative mechanism design for a prosthetic hand Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medi cine Vol. 218 No. 6. p p. 451 459.

PAGE 75

63 [21] Crowder, R.M. and Dubey, V.N. (2004). Grasping and Control Issues in Adaptive End Efforts ASME 2004 Design Engineering Technical Conferences and Computers and Information in Engineering Conference. [22] Dolla r A M. and Howe R. D. ( 2006 ). Joint Coupling design of underactuated grippers. Proceedings of IDETC/CIE 2006 ASME 2006 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference September 10 13. Philade lphia, Pennsylvania, USA [23] Bundhoo, V. and Park, E.J., Design of an Artificial Muscle Actuated Finger towards Biomimetic Prosthetic Hands. [24] Howell, L.L. (2001). Compliant Mechanisms John Wiley & Sons, Inc. New York. [25] Napier, J.R. (1956) The prehensile movements of the human hand The Journal of Bone and Joint Surgery. Vol. 38B. No. 4. pp. 902 913. [26] Sharp, G. and Thompson, D., Biomechanics of the hand [27] Kamakura, N., Matsuo, M., Ishii, H., Mitsuboshi, F., and Miura, Y. (1980) Patterns of static prehension in normal hands The American Journal of Occupational Therapy. Vol. 34. pp. 437 445. [28] Light C M. (2000). An intelligent hand prosthesis and evaluation of pathological and prosthetic hand function [dissertation]. Sou thampton (UK). Uni versity of Southampton. pp. 179 91. [29] Light, C.M., Ch appell, P.H., and Kyberd, P.J. (2002). Establishing a Standardized Clinical Assessment Tool of Pathologic and Prosthetic Hand Function: Normative Data, Reliability, and Validity Archieves of Physical Medicine and Rehabilitation Vol. 83. pp.777 783. [30 ] Stein, R.B., and Walley, M. (1983). Functional comparison of upper extremity amputees using myoelectric and conventional prosthesis Archieves of Physical Medicine and Rehabilit ation. Vol 64. June. pp. 243 278. [3 1 ] Alqasemi, R., Mahler, S. and Dubey, R. (2007). A Double Claw Robotic End Effector Design Florida Conference on Recent Advances in Robotics. May 31 June 1.

PAGE 76

64 [32] Weeks, D.L. (2003). The role of variability in pra ctice structure when learning to use an upper extremity prosthesis. Journal of P rosthetics and Orthotics. Vol. 15. Num. 3. pp 84.

PAGE 77

65 Appendices

PAGE 78

66 Appendix A: Instantaneous stability of the crossed four bar m echanism Figure A .1 : Instantaneous stability of crossed four bar mechanism for a=0 Figure A .2: Instanta neous stability of crossed four bar mechanism for a=0.1

PAGE 79

67 Appendix A: (Continued) Figure A .3 : Instanta neous stability of crossed four bar mechanism for a=0.2 Figure A .4 : Instanta neous stability of crossed four bar mechanism for a=0.3

PAGE 80

68 Appendix A: (Continued) Figure A .5 : Instantaneous stability of crossed four bar mechanism for a=0.4 Figure A .6 : Instanta neous stability of crossed four bar mechani sm for a=0.5

PAGE 81

69 Appendix A: (Continued) Figure A .7 : Instanta neous stability of crossed four bar mechanism for a=0.6 Figure A .8 : Instanta neous stability of crossed four bar mechanism for a=0.7

PAGE 82

70 Appendix A: (Continued) Figure A .9 : Instantaneous stability of crossed four bar mechanism for a=0.8 Figure A .10 : Instantaneous stability of crossed four bar mechanism for a=0.9

PAGE 83

71 Appendix A: (Continued) Figure A .11 : Instanta neous stability of crossed four bar mechanism for a=1.0

PAGE 84

72 Appendix B: SHAP Test Protocol by C olin Light [28]. B.1 Setting up the assessment The subject should be seated at a table. With relaxed shoulders and arms resting on the 0 angle. Place the test platform (red/blue sided) directly in front of the subject (blue side facing upwards), approximately 3 inches from the front edge of the table. Fit the timer unit into the pace provided in the front of the platform. For each of the following abstract tasks, the board should be moved from left to right so that each task is directly in front of the subject, thereby ensuring no bias towards one hand. The case and all ADL objects may be removed from the table. B.2 Procedural notes Each task should be demonstrated to the subject using slow, clear movements ensuring that the subject is aware of the appropriate grip. The subject should be given the opportunity to ask questions prior to the commencement of each task. It is important to note that the demonstration should be carried out using the corresponding hand under assessment, to avoid any confusion for the subject. Prosthesis users should be encouraged to practice each task, prior to timing the event, in order to determine the most appropriate technique (as many users often carry out tasks with the natu ral hand alone). Due to the difficulties associated with myoelectric prostheses, if it is apparent that the device has failed to respond to user demand, than a

PAGE 85

73 Appendix B : (Continued) note should be made, and a retest allowed. If the prosthesis is simil arly unresponsive, the second task time should be recorded and a note made of the difficulties encountered. Only one chance to carry out the timed task should be given, unless a serious handling error causes an unrealistic result. The time to complete the task (and the appropriate grip if readily identifiable) should be recorded, as well as any relevant notes. When establishing any form of normative data it is imperative that the task is carried out fully. Due to the need to complete in the minimum time, there is frequently a these circumstances the task should be repeated. B.3 Abstract o bjects The lightweight objects are to be used first. If a subject cannot complete t he task, this should be recorded as C/C (cannot complete). The task involves moving the object from the rear slot to the front slot. Only the hand under assessment should be used for any of these 1. barrier. Move the board so t hat these slots are directly in front of the subject (maintaining the distance from the slot to the table). Using a spherical grip, move the ball over the barrier to the front slot.

PAGE 86

74 Appendix B : (Continued) Start the timer, pick up and move the object a s demonstrated with as few mistakes as possible, and as quickly as possible, to the front slot. Complete the task by depressing the 2. move the object to the front slot. 3. that these slots are directly in fro nt of the subject (maintaining the distance from the front of the table). Using the power grip, pick up the object by the cylinder (between the two markers), and move to the front slot. Start the timer, pick up the object between the two markers as demon strated, and 4. towards the subject. Move the board so that these slots are directly in front of th e subject (maintaining the distance from the front of the table). Using a lateral grip, pick up the object by the handle, and move to the front slot. the front slot, and then 5. grip, move the object to the front slot.

PAGE 87

75 Appendix B : (Continued) the f 6. extension grip (with the thumb in front of the object, and fingers extended flat on the rear side), move the object to the front slot. rt the timer, move the object as demonstrated and as quickly as possible to The procedure should now be repeated, in the same order using the metal objects. If a subject has failed to complete tasks lightweight obj ects, then the appropriate heavier object tasks may be ignored (to avoid undue strain on the subject). In this instance, a Once completed, place the form board objects in the foam. Turn the test p latform over (the red side facing upwards), and position as before, with the timer on the place provided. The platform should remain centred in front of the subject for all ADL tasks. B.4 Activities of d aily l iving Each task should be demonstrated to the subject using slow, clear movements, ensuring that the subject is aware of the appropriate grip. of this assessment consists of 14 everyd ay activities, which should be timed in the same manner by depressing the blue button to start and stop the

PAGE 88

76 Appendix B : (Continued) timer. Again tasks should be completed as quickly as possible, with as few mistak es as possible, using 1. Pick up coins: arrange the two 2p and two 1p coins in the designated areas on the red platform. Place the jar in the designated spot for this test with the lid removed. Pick up e ach coin in turn (by sliding to the edge of the platform), using a tip or tripod grip, and drop into the jar. Move from right to left. Reset the task. lift each coin in turn as quickly a s possible drop in the jar, as demonstrated and 2. Button board: Place the button board to the right of the timer unit if assessing the right hand, and to the left if assessing the left hand. The buttons should be farthest from the timer units. Undo each button in turn, using only the assessed hand (as a test of dexterity) in a tripod grip. The other hand may be used to steady the board, but may not assist in the task. The board should remain on the platform. Reset the task. and using only the appropriate hand, undo all four buttons in any order as demonstrated and as quickly as possible You may steady the board with your other hand so that it remains on the platform throughout the task. Then stop the timer using only the appropriate hand. You may now practice this task. 3. Cutting: place the knife to the side of the timer unit (approximately arranged for red platform. Pick up the knife and using the other hand to steady the object, cut it

PAGE 89

77 Appendix B : (Continued) clearly into two sections. Then replace the knife on the platform, remould the plasticine, and reset the task. use the knife provided to cut the plasticine object clearly into two sections, as demonstrated an d as quickly as possible You may use the other hand to steady the object. Return the knife to the platform, 4. Simulated page turning: place the 4 inch by 6 inch card in the designated area on the opposing side of the platform to the hand under assessment. Using an extension or tripod grip, pick up the card, turn over and place in the opposite designated area (as if turning the page of a book). Reset the task. lift, turn over (as if turning the page of a book), and replace the card on the platform, as demonstrated and as quickly as possible T hen stop the 5. Jar lid: the lid should be placed on the empty jar, and tightened only with sufficient force as would be expected for everyday use/shelf storage. The ja r should be placed in the designated area on the red platform. Both hands should be used for this task. Pick up the jar with the non assessed hand, undo the lid, (using a flexion grip with the lid firmly in the palm to form a combined power/precision grip) using the assessed hand, and return both the jar and the lid to the platform. Reset the task. pick up the jar, and undo the lid with the hand under assessment a s demonstrated and as quickly as possible Return the jar and the lid to the platform and

PAGE 90

78 Appendix B : (Continued) To avoid repetitive filling/emptying of objects with water during the following four tasks, it is advisable to fill a separate container with approximately one pint of water. It may also be advisable to have a towel nearby. 6. Pouring from jug: fill the jug with 100 ml of water (100 ml is marked on the jug). Place the jug on the designated area on the red test platform, with the handle pointing to the right for right handed subjects, and to the left for lef t handed subjects. Place the jar (without lid) on the designated left area for right handed people, and on the designated right area for the left handed people. Lift the jug by the handle (in a lateral grip), and pour the water into the jar. Reset the task and whilst ensuring as little spillage as possible, pour the water from the jug to the jar, as demonstrated and as quickly as possible T hen stop the timer. You should avoid trying to empty the jug to every last drop, and merely ensur e the vast majority of the water has been transferred. 7. Pouring from carton: fill the carton with 200 ml of water. Place in the designated area on the red platform with the spout pointing towards the jar (according to the handedness criteria described for the previous test). Pick up the carton using a Reset the task. and whilst ensuring as little spillage as possible, pour the water from the carton to the jar, as demonstrated and as quickly as possible T hen stop the timer. Again you should avoid trying to empty the jug to every last drop, and merely ensure the vast majority of the water has been transferred.

PAGE 91

79 Appendix B : (Continued) 8. Large heavy object: fi ll the jar with water (to the full mark), and tighten the lid. Place in the designated area on the left side of the red platform (for right handed), or on the right side (for left handed people). Place the empty carton lengthways along the middle of the pl atform (without obscuring the timer) to create a barrier. Lift the jar over the carton, using a power grip, and place in the opposing marked area. move the jar over the carton to the opposing marked area, as demonstrated and as quickly as possible T The water may now be disposed of and any will form no further part of the assessment procedures. 9. Large light object: place the empty tin in the appropriate area on the left hand side of the red platform (if right handed), or on the right hand side (if left handed). Place the carton to create a barrier as before. Lift the tin over the carton, using a power grip, and place on the opposing marked area. move the tin over the carton to the opposing marked ar ea, as demonstrated and as quickly as possible T Place the test unit (with foam inside) on the table, directly in front of the subject, 3 inches from the front. Place the platform on the foam base and the timer unit on the appropriat e slot. The final 5 tasks will involve the use of the unit.

PAGE 92

80 Appendix B : (Continued) 10. Lift tray: place the platform (red side upwards), on the table to the left of the test unit (for right handers), or to the right (for left handers), with the board slight ly overhanging the front of the table by approximately one inch, with the long edge facing forwards. The timer should remain in the unit. Both hands should be used to pick up the platform, using a lateral (or extension grip). Assuming a right hander: lift on the table to the right of the unit. Return the platform to the left hand side of the unit. tray from left to right hand side of the test unit, as demonst rated and as quickly as possible T 11. Rotate key: return the platform to the test unit base (red side upwards). Place the key on the lock so it appears vertical. Turn the key to the white mark using a lateral grip. rot ate the key as demonstrated and as quickly as possible at least one quarter turn clockwise, to the white mark, and release (at which time the key will spring back), 12. Open/close zip: ensure the zip is closed and lies flat against the back board. Open and close the zip using a lateral, or two point tip grip. Start the time open and then close the zipper in as short as time as possible, as 13. Rotate screw: place the screwdriver in the designat ed area on the red platform (on the right hand side for a right handed subject, or on the left for a left handed

PAGE 93

81 Appendix B : (Continued) subject). The screw is mounted on a clip, which should be attached to the front of the case. Use the area directly in front of the screwdriver (between the handle and clasp on the case). Ensure the arrow is vertical. Use two hands to guide the screwdriver to the screw, and rotate it 90 clockwise to the mark using one hand only (in a combined power/precision grip, also kno wn as a diagonal volar grip). Reset the task. use the screwdriver to rotate the screw a quarter turn clockwise to, or beyond, the white mark, as demonstrated and as quickly as possible Once completed, the screwdriver should be replaced on the platform and the timer stopped. Two hands may be used to guide the screwdriver to the screw, but only the appropriate hand should be used in turning the screwdriver. 14. Door handle: rotate the door handle (using a hook or power grip) until it is full y open, and then release. rotate the door handle until it is fully open, and then release, as demonstrated and as quickly as possible. T

PAGE 94

82 Table C.1: Lightweight abstract objects time for subject #1 A natomical h and (sec) H ook (sec) H ook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Ave rage Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Spherical 1.28 0.90 0.89 1.02 3.55 3.46 3.39 3.47 2.93 2.83 2.88 2.88 Tripod 1.19 1.13 1.37 1.23 3.38 3 .77 3.84 3.66 2.67 2.93 3.58 3.06 Power 1.15 1.14 1.28 1.19 3.61 2.49 2.40 2.83 2.69 2.94 2.53 2.72 Lateral 1.35 1.39 1.41 1.38 3.04 3.07 2.86 2.99 3.22 3.03 2.60 2.95 Tip 1.39 1.32 1.41 1.37 3.70 3.23 3.76 3.56 2.70 3.23 2.93 2.95 Extension 1.45 1.71 1.63 1.60 3.05 3.53 2.73 3.10 2.99 3.83 3.94 3.59 Table C.2: Heavyweight abstract objects time for subject #1 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 0.97 1.05 0.93 0.98 4.49 4.82 5.31 4.87 4.11 2.83 2.68 3.21 Tripod 1.27 1.14 1.25 1.22 3.55 3.20 2.97 3.24 2.76 2.86 2.94 2.85 Power 1.13 0.99 0.95 1.02 3.59 5.34 4.28 4.40 2.74 4.30 2.77 3.27 Lateral 1.30 1.34 1.23 1. 29 3.72 3.78 2.63 3.38 3.04 2.89 2.73 2.89 Tip 1.10 1.17 1.39 1.22 3.51 5.46 4.36 4.44 3.43 1.87 3.05 2.78 Extension 2.16 1.99 1.83 1.99 3.67 4.60 4.97 4.41 3.18 3.28 3.75 3.40 Appendix C: Human testing data

PAGE 95

83 Table C .3 : Activities of daily living time data for subject #1 Anato mical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 4.01 3.44 3.73 3.73 17.95 13.09 12.37 14.47 29.02 15.82 15.49 20.11 Undo buttons 7.1 1 4.42 6.58 6.04 15.54 19.18 17.72 17.48 0.00 0.00 0.00 0.00 Food cutting 3.05 3.17 2.41 2.88 0.00 0.00 0.00 0.00 24.47 0.00 0.00 0.00 Page turning 1.70 1.54 1.51 1.58 5.08 4.90 5.24 5.07 5.93 4.90 5.33 5.39 Jar lid 1.65 1.57 1.77 1.66 32.40 11.48 21.18 21.69 7.23 4.87 6.93 6.34 Jug pour 3.69 3.57 3.72 3.66 12.13 9.68 11.84 11.22 15.09 8.55 8.34 10.66 Carton pour 7.98 7.33 7.25 7.52 14.69 13.29 13.80 13.93 14.98 14.70 13.55 14.41 Full jar 1.53 1.45 1.46 1.48 5.32 4.94 5.46 5.24 4.39 3.37 4.02 3.93 E mpty tin 1.55 1.41 1.55 1.50 4.17 4.57 3.93 4.22 3.17 3.49 3.11 3.26 Tray 2.93 2.77 2.74 2.81 9.57 9.19 9.82 9.53 9.62 9.60 7.95 9.06 Turn a key 1.32 1.17 1.25 1.25 3.43 3.47 3.44 3.45 3.57 3.59 3.55 3.57 Zipper 2.26 2.09 2.05 2.13 7.07 5.45 4.69 5.74 11.99 6.72 6.46 8.39 Screw 3.27 2.90 3.19 3.12 10.23 10.63 8.12 9.66 8.87 9.44 8.03 8.78 Door handle 1.47 1.40 1.35 1.41 2.78 3.22 3.12 3.04 2.97 2.97 2.72 2.89 App endix C: (Continued)

PAGE 96

84 Table C .4 : Lightweight a bstract o bjects time for s ubject #2 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 2.33 2.59 1.75 2.22 4.04 4.57 4.21 4.27 4.28 4.13 4.65 4.35 Tripod 2.38 2.02 2.19 2.20 5.07 4.43 4.80 4 .77 3.82 4.22 5.22 4.42 Power 2.01 1.92 2.00 1.98 4.10 4.31 3.73 4.05 3.68 3.25 3.49 3.47 Lateral 2.17 2.14 1.83 2.05 4.90 4.86 3.99 4.58 5.39 4.91 6.13 5.48 Tip 1.96 2.00 1.74 1.90 5.57 5.23 4.36 5.05 4.02 4.29 4.84 4.38 Extension 2.20 2.42 2.22 2.28 5.05 5.44 5.04 5.18 4.71 4.61 5.03 4.78 Table C.5 : Heavyweight a bstract o bjects time for s ubject #2 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Tria l 3 Average Spherical 2.26 2.40 2.49 2.38 4.35 4.20 5.15 4.57 4.72 4.11 4.23 4.35 Tripod 1.62 1.83 1.93 1.79 3.21 3.45 4.18 3.61 3.54 3.17 3.01 3.24 Power 1.89 2.23 1.99 2.04 5.23 3.89 4.15 4.42 3.89 2.93 3.97 3.60 Lateral 2.13 1.89 2.23 2.08 4.36 4.39 4.28 4.34 3.97 3.84 3.08 3.63 Tip 1.80 1.82 2.24 1.95 3.44 3.27 4.03 3.58 3.07 3.10 3.02 3.06 Extension 2.40 2.04 2.09 2.18 5.39 5.28 5.69 5.45 5.26 5.34 4.85 5.15 Appendix C: (Continued)

PAGE 97

85 Table C .6 : Activities of daily living time for subject #2 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 6.37 5.42 5.37 5.72 19.42 19.29 15.30 18.00 29.96 22.59 22.14 24.90 Undo buttons 8.77 6.53 5.87 7.06 25.69 1 7.09 28.30 23.69 24.48 32.73 28.97 28.73 Food cutting 5.07 5.42 5.33 5.27 32.11 37.01 19.50 29.54 0.00 0.00 0.00 0.00 Page turning 2.68 2.65 2.65 2.66 9.09 6.68 4.72 6.83 5.68 6.14 5.41 5.74 Jar lid 3.19 2.70 3.14 3.01 0.00 0.00 0.00 0.00 5.95 6.20 5.03 5.73 Jug pour 5.28 5.16 5.24 5.23 16.52 11.34 11.77 13.21 13.45 10.52 9.22 11.06 Carton pour 8.59 8.25 9.12 8.65 14.87 14.93 16.21 15.34 14.37 14.28 12.67 13.77 Full jar 2.84 3.09 2.97 2.97 6.68 6.22 6.72 6.54 6.38 7.15 4.91 6.15 Empty tin 2.30 2.35 2.37 2.34 5.06 5.78 5.60 5.48 4.59 3.45 3.50 3.85 Tray 4.29 3.97 3.57 3.94 10.65 11.23 9.89 10.59 8.16 8.42 8.53 8.37 Turn a key 1.62 1.80 1.84 1.75 5.84 4.37 4.97 5.06 3.78 3.67 3.22 3.56 Zipper 4.47 4.50 3.13 4.03 10.57 13.22 11.29 11.69 0.00 0.00 0. 00 0.00 Screw 4.69 4.69 4.13 4.50 12.40 10.30 11.15 11.28 11.52 7.83 9.85 9.73 Door handle 1.97 1.92 2.09 1.99 5.20 5.38 5.59 5.39 4.63 3.59 3.89 4.04 Appendix C: (Continued)

PAGE 98

86 Table C.7: Lightweight abstract objects time for subject #3 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 3.09 2.49 2.14 2.57 5.16 3.74 3.78 4.23 4.69 4.25 3.19 4.04 Tripod 1.71 1.59 1.79 1.70 4.63 3.89 4.62 4.38 3.82 3. 85 3.83 3.83 Power 1.62 1.59 1.63 1.61 3.73 3.87 3.67 3.76 3.54 3.47 4.12 3.71 Lateral 2.27 2.17 1.80 2.08 4.62 4.18 4.99 4.60 5.23 5.07 5.24 5.18 Tip 2.19 2.16 1.98 2.11 5.04 4.26 5.57 4.96 4.57 5.17 4.52 4.75 Extension 1.90 1.97 2.21 2.03 4.97 4.99 5 .81 5.26 5.37 4.24 6.07 5.23 Table C.8: Heavyweight abstract objects time for subject #3 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Avera ge Spherical 1.89 2.07 1.94 1.97 0.00 0.00 0.00 0.00 3.73 4.49 4.61 4.28 Tripod 1.38 1.55 1.33 1.42 4.58 3.85 3.52 3.98 3.73 3.60 3.51 3.61 Power 1.67 1.76 1.77 1.73 4.06 4.78 4.35 4.40 4.03 3.64 3.63 3.77 Lateral 1.89 1.62 1.73 1.75 5.17 3.94 3.93 4.3 5 5.55 4.77 4.61 4.98 Tip 1.63 1.67 1.55 1.62 7.05 5.83 5.14 6.01 5.91 5.11 4.88 5.30 Extension 2.37 1.74 2.03 2.05 6.29 6.80 4.97 6.02 6.49 4.28 4.49 5.09 Appendix C: (Continued)

PAGE 99

87 Table C .9 : Activities of daily living time data for subject #3 A natomical hand (sec) H oo k (sec) H ook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage C oins 5.29 5.53 5.06 5.29 19.87 15.34 16.81 17.34 36.40 26.03 21.53 27.99 Undo buttons 6.75 6.69 4.87 6.10 25.33 19.6 8 12.93 19.31 0.00 0.00 0.00 0.00 F ood cutting 5.72 5.29 4.09 5.03 52.52 31.47 34.80 39.60 0.00 0.00 0.00 0.00 P age turning 1.99 2.15 2.62 2.25 7.47 6.43 8.21 7.37 9.32 10.69 8.83 9.61 J ar lid 3.07 2.19 2.61 2.62 0.00 0.00 0.00 0.00 11.54 8.28 8.24 9.35 J ug pour 5.35 4.77 3.82 4.65 14.28 9.59 7.21 10.36 11.91 15.91 7.32 11.71 C arton pour 8.92 8.31 7.69 8.31 18.27 21.07 18.28 19.21 16.47 14.09 16.50 15.69 Full jar 2.12 2.29 2.31 2.24 6.93 4.70 5.53 5.72 4.76 4.53 5.88 5.06 E mpty tin 1.80 1.93 1.91 1. 88 3.99 3.87 3.67 3.84 4.07 4.09 3.77 3.98 T ray 3.42 3.64 3.33 3.46 11.58 8.09 8.94 9.54 9.35 9.88 11.28 10.17 Turn a key 1.96 1.61 1.65 1.74 5.21 4.14 3.97 4.44 6.16 5.77 6.73 6.22 Z ipper 2.64 2.79 2.75 2.73 7.28 20.56 13.33 13.72 22.36 20.13 20.99 21 .16 S crew 3.88 4.33 4.00 4.07 18.34 11.27 8.02 12.54 13.52 10.04 10.48 11.35 D oor handle 1.99 1.88 2.12 2.00 3.28 3.42 3.53 3.41 4.03 3.63 3.00 3.55 Appendix C: (Continued)

PAGE 100

88 Table C .10 : Lightwe ight abstract o bjects time for s ubject #4 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 2.11 1.92 1.97 2.00 4.55 4.74 4.26 4.52 4.18 4.19 4.74 4.37 Tripod 2.18 2.49 1.93 2.20 5.18 4.33 5.11 4.87 3.21 2.9 9 3.30 3.17 Power 2.19 2.15 1.95 2.10 4.85 4.03 4.55 4.48 3.48 3.59 3.97 3.68 Lateral 2.01 2.30 1.75 2.02 5.17 4.38 6.59 5.38 4.78 3.67 3.69 4.05 Tip 1.91 1.83 1.90 1.88 5.11 5.75 5.87 5.58 5.14 4.11 3.59 4.28 Extension 2.53 2.83 2.65 2.67 5.32 4.24 4. 20 4.59 4.47 3.83 3.78 4.03 Table C.11: Heavyweight abstract objects time for subject #4 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Averag e Spherical 1.96 2.02 2.14 2.04 5.11 4.31 4.01 4.48 4.84 4.14 4.02 4.33 Tripod 1.93 1.73 1.97 1.88 4.29 4.73 5.08 4.70 3.26 3.24 2.97 3.16 Power 2.19 2.04 2.01 2.08 6.34 5.33 5.07 5.58 4.49 4.76 4.14 4.46 Lateral 2.03 2.39 2.04 2.15 5.31 4.63 4.90 4.95 4.03 3.80 5.23 4.35 Tip 2.09 1.88 2.29 2.09 7.27 6.71 7.42 7.13 6.77 5.70 5.78 6.08 Extension 2.04 2.02 2.42 2.16 5.48 5.32 4.93 5.24 4.45 4.32 4.61 4.46 Appendix C: (Continued)

PAGE 101

89 Table C .12 : Activ ities of daily living time data for s ubject #4 Anatomical hand (sec) Hoo k (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 5.02 5.08 4.96 5.02 23.97 15.61 15.18 18.25 17.52 17.39 15.47 16.79 Undo buttons 8.56 8.13 7.30 8.00 27.54 101. 46 123.23 84.08 41.58 48.66 51.04 47.09 Food cutting 7.07 4.93 4.79 5.60 24.92 43.49 26.63 31.68 30.13 27.44 18.71 25.43 Page turning 2.63 2.34 1.83 2.27 8.03 4.59 5.61 6.08 4.99 4.44 4.98 4.80 Jar lid 3.13 2.65 2.73 2.84 16.33 7.11 8.61 10.68 6.54 7.00 5.27 6.27 Jug pour 7.07 6.20 5.65 6.31 16.64 12.30 12.67 13.87 14.37 12.70 9.89 12.32 Carton pour 9.51 9.82 8.44 9.26 15.27 16.20 15.00 15.49 17.99 17.83 18.33 18.05 Full jar 2.40 2.14 2.34 2.29 5.30 4.45 4.53 4.76 5.77 5.97 4.53 5.42 Empty tin 1.93 2.03 1.87 1.94 3.69 3.83 3.66 3.73 3.52 3.20 3.29 3.34 Tray 3.83 3.69 4.03 3.85 14.71 11.92 10.29 12.31 12.28 9.17 8.63 10.03 Turn a key 2.07 1.93 1.69 1.90 5.83 4.80 4.39 5.01 4.28 4.11 4.33 4.24 Zipper 3.25 2.71 2.92 2.96 7.66 16.73 10.77 11.72 14.40 10.14 7.99 10.84 Screw 4.61 4.17 4.06 4.28 13.99 12.44 17.07 14.50 10.72 9.20 8.38 9.43 Door handle 2.07 1.97 1.92 1.99 2.88 4.05 3.12 3.35 3.80 3.82 3.55 3.72 Appendix C: (Continued)

PAGE 102

90 Table C .13: Lightweight abstract objects time for s ubject #5 Anatomical hand ( sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 1.43 1.33 1.11 1.29 3.13 2.93 2.68 2.91 3.90 3.50 3.63 3.68 Tripod 1.63 1.26 1.13 1.34 3.80 3.27 2 .85 3.31 4.63 3.76 3.30 3.90 Power 1.18 1.11 1.09 1.13 2.99 2.69 2.68 2.79 4.05 3.36 2.73 3.38 Lateral 1.49 1.83 1.46 1.59 3.73 3.59 3.47 3.60 3.92 3.88 3.95 3.92 Tip 1.41 1.41 1.03 1.28 3.67 3.85 4.07 3.86 3.42 3.90 3.65 3.66 Extension 1.64 1.90 1.94 1.83 5.07 3.99 4.23 4.43 3.50 3.16 3.05 3.24 Table C .14: Heavyweight abstract o bjects time for s ubject #5 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 1.27 1.29 1.28 1.28 5.93 3.90 3.65 4.49 5.21 4.69 3.43 4.44 Tripod 1.28 1.09 1.07 1.15 3.07 2.73 2.61 2.80 3.29 3.33 3.19 3.27 Power 1.50 1.55 1.63 1.56 4.59 4.63 3.87 4.36 3.33 3.10 2.97 3.13 Lateral 1.83 1.53 1.43 1.60 4.0 2 4.07 4.02 4.04 3.72 3.79 4.11 3.87 Tip 1.20 1.11 1.29 1.20 3.32 3.69 3.44 3.48 4.18 3.17 3.59 3.65 Extension 1.64 1.57 1.77 1.66 6.23 5.93 5.54 5.90 4.81 3.69 3.50 4.00 Appendix C: (Continued)

PAGE 103

91 Table C. 15 : Activities of daily living time data for subject #5 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 4.69 4.29 3.66 4.21 16.68 20.87 15.23 17.59 12.17 15.27 11.03 12.82 Undo buttons 5.96 4.77 4.02 4.92 22.57 15.99 31.10 23.22 40.37 31.79 28.37 33.51 Food cutting 3.76 3.69 3.53 3.66 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Page turning 1.74 1.69 1.46 1.63 8.42 4.76 4.48 5.89 3.97 3.78 3.17 3.64 Jar lid 2.18 2.11 2.05 2.11 22.19 8.73 11.23 14.05 4.6 9 4.56 3.01 4.09 Jug pour 4.27 4.33 3.79 4.13 17.23 11.10 13.53 13.95 6.83 6.97 7.42 7.07 Carton pour 8.93 8.93 3.79 7.22 12.98 11.26 11.42 11.89 12.51 15.09 12.14 13.25 Full jar 2.14 1.73 1.88 1.92 4.33 4.97 3.97 4.42 4.13 3.77 3.36 3.75 Empty tin 1. 56 1.57 1.51 1.55 3.56 3.32 3.00 3.29 3.61 3.67 3.15 3.48 Tray 3.75 2.59 2.67 3.00 8.35 11.36 9.61 9.77 8.33 8.09 7.09 7.84 Turn a key 1.47 1.39 1.58 1.48 3.84 3.59 3.10 3.51 3.23 2.79 2.48 2.83 Zipper 2.17 1.67 1.62 1.82 13.51 6.83 4.23 8.19 7.39 8.35 4.78 6.84 Screw 3.78 4.57 3.30 3.88 6.45 4.92 8.30 6.56 7.83 5.99 6.97 6.93 Door handle 1.19 1.12 1.23 1.18 3.72 3.39 3.45 3.52 3.17 2.83 2.54 2.85 Appendix C: (Continued)

PAGE 104

92 Table C .16: Lightweight a bstract o bjects time for subject #6 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 0.99 0.83 1.03 0.95 4.32 4.36 3.19 3.96 3.02 2.53 2.63 2.73 Tripod 0.87 0.87 0.92 0.89 3.61 2.64 2.52 2.92 3.37 2.9 0 3.37 3.21 Power 1.20 1.15 1.12 1.16 3.32 3.07 2.83 3.07 2.80 2.51 2.51 2.61 Lateral 1.20 1.24 1.12 1.19 3.69 3.21 2.67 3.19 3.99 4.27 4.21 4.16 Tip 1.76 1.34 1.26 1.45 4.25 3.64 3.24 3.71 3.77 2.93 2.69 3.13 Extension 1.83 2.00 1.63 1.82 4.05 2.83 3. 39 3.42 3.20 3.18 2.64 3.01 Table C.17: Heavyweight abstract objects time for subject #6 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Aver age Spherical 1.09 1.14 1.07 1.10 4.78 5.73 5.55 5.35 2.87 3.08 2.89 2.95 Tripod 0.95 0.79 0.98 0.91 3.43 3.21 2.65 3.10 3.20 3.40 3.33 3.31 Power 1.12 1.13 1.11 1.12 11.88 5.23 7.61 8.24 3.08 3.10 2.77 2.98 Lateral 1.42 1.20 1.26 1.29 4.40 4.51 3.77 4 .23 4.09 4.35 3.69 4.04 Tip 1.53 1.27 1.26 1.35 3.99 4.14 3.77 3.97 3.43 3.86 3.35 3.55 Extension 1.67 1.67 1.74 1.69 3.92 3.60 3.71 3.74 2.76 3.20 2.90 2.95 Appendix C: (Continued)

PAGE 105

93 Table C. 18 : Activities of daily living time data for subject #6 Anatomical hand (sec ) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 3.94 3.89 4.03 3.95 18.33 20.14 24.92 21.13 17.69 18.40 17.79 17.96 Undo buttons 6.97 6.25 6.57 6.60 32.2 3 22.12 22.56 25.64 44.99 21.80 22.37 29.72 Food cutting 3.75 2.75 3.60 3.37 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Page turning 1.23 1.30 1.69 1.41 4.77 3.64 4.57 4.33 5.87 5.24 5.53 5.55 Jar lid 1.74 1.81 1.67 1.74 5.76 8.21 8.37 7.45 5.89 4.97 4.99 5.28 Jug pour 3.80 5.27 4.02 4.36 17.01 12.99 13.00 14.33 9.34 7.85 9.15 8.78 Carton pour 7.95 8.68 7.94 8.19 41.47 16.93 17.37 25.26 17.16 16.59 16.53 16.76 Full jar 1.93 1.87 1.69 1.83 6.38 4.97 4.96 5.44 3.35 4.14 3.93 3.81 Empty tin 1.54 1.42 1.22 1.39 5.28 4.97 5.29 5.18 2.77 2.63 2.84 2.75 Tray 2.87 2.91 2.97 2.92 10.21 10.65 8.88 9.91 11.89 9.17 9.72 10.26 Turn a key 1.07 0.77 0.89 0.91 4.17 4.01 3.38 3.85 4.35 4.59 3.86 4.27 Zipper 1.68 1.97 1.83 1.83 9.58 7.70 8.21 8.50 0.00 0.00 0.00 0.00 Screw 3.45 3.00 2.84 3.10 6.29 10.18 7.13 7.87 7.68 6.88 7.29 7.28 Door handle 1.15 1.01 0.89 1.02 1.59 1.37 1.24 1.40 2.49 2.44 2.22 2.38 Appendix C: (Continued)

PAGE 106

94 Table C. 19: Lightweight abstract o bjects time for s ubject #7 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 1.25 1.17 1.39 1.27 6.19 4.88 4.87 5.31 5.93 4.88 4.65 5.15 Tripod 1.15 1.12 1.07 1.11 5.61 4.09 3.69 4.46 5.11 7.67 4.59 5. 79 Power 1.37 1.31 1.23 1.30 4.47 4.95 5.87 5.10 4.93 6.60 7.08 6.20 Lateral 1.53 1.50 1.84 1.62 5.73 4.57 4.44 4.91 9.31 8.24 4.88 7.48 Tip 1.37 1.36 1.15 1.29 6.41 5.93 4.36 5.57 4.66 5.02 5.63 5.10 Extension 1.67 1.56 1.17 1.47 7.57 5.47 4.13 5.72 7 .43 7.32 8.37 7.71 Table C.20: Heavyweight abstract objects time for subject #7 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spher ical 1.37 1.19 1.30 1.29 5.42 4.47 4.54 4.81 5.57 9.83 8.53 7.98 Tripod 1.08 1.02 1.06 1.05 5.05 5.41 4.81 5.09 5.03 3.67 7.20 5.30 Power 1.30 1.20 1.33 1.28 11.82 0.00 0.00 3.94 7.06 5.69 6.12 6.29 Lateral 1.57 1.53 1.57 1.56 5.09 4.94 3.25 4.43 7.12 5 .73 8.15 7.00 Tip 1.36 1.20 1.42 1.33 4.70 4.45 3.68 4.28 7.03 5.89 10.02 7.65 Extension 1.57 1.62 1.60 1.60 6.91 4.97 4.54 5.47 6.87 4.21 4.93 5.34 Appendix C: (Continued)

PAGE 107

95 Table C.21: Activities of daily living time data for subject #7 Anatomical hand (sec) Hook (sec ) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 5.99 5.02 4.29 5.10 18.85 16.17 15.14 16.72 35.92 18.55 16.65 23.71 Undo buttons 9.06 6.86 5.08 7.00 20.61 33.53 15.2 2 23.12 0.00 0.00 0.00 0.00 Food cutting 6.75 3.89 6.78 5.81 26.93 21.45 18.13 22.17 24.53 30.92 20.86 25.44 Page turning 2.03 1.77 1.65 1.82 6.54 5.29 5.34 5.72 8.42 7.55 5.69 7.22 Jar lid 1.89 2.24 1.85 1.99 19.60 7.79 6.64 11.34 8.65 16.87 7.34 10.95 Jug pour 6.60 5.43 5.31 5.78 10.69 9.8 11.54 10.68 10.83 9.48 8.97 9.76 Carton pour 8.60 9.95 9.21 9.25 17.13 18.48 13.9 16.50 15.26 14.71 10.26 13.41 Full jar 1.94 1.89 1.46 1.76 4.68 4.16 3.33 4.06 6.03 8.57 6.8 7.13 Empty tin 1.18 1.21 1.24 1.21 2 .88 2.34 2.79 2.67 6.04 3.67 3.05 4.25 Tray 3.09 3.68 2.97 3.25 10.87 11.36 6.65 9.63 16.53 10.59 8.79 11.97 Turn a key 1.24 1.38 1.43 1.35 3.61 2.97 2.67 3.08 3.97 3.49 3.00 3.49 Zipper 2.18 1.86 1.77 1.94 17.49 9.47 13.6 13.52 14.83 5.37 5.94 8.71 S crew 3.68 3.07 4.03 3.59 17.63 10.59 8.19 12.14 10.94 9.26 8.63 9.61 Door handle 1.32 1.32 1.19 1.28 2.09 1.67 1.59 1.78 2.57 3.13 2.59 2.76 Appendix C: (Continued)

PAGE 108

96 Table C.22: Lightweight abstract o bjects time for s ubject #8 Anatomical hand (sec) Hook (sec) Hook wit h fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 1.53 1.73 1.43 1.56 3.41 3.38 3.08 3.29 3.02 3.26 2.73 3.00 Tripod 1.93 1.78 1.60 1.77 3.81 3.00 3.43 3.41 2.84 2.69 2.42 2 .65 Power 1.39 1.26 1.34 1.33 2.63 2.88 2.94 2.82 2.69 2.65 2.34 2.56 Lateral 1.66 1.58 1.33 1.52 3.44 2.58 2.37 2.80 2.46 3.00 2.15 2.54 Tip 1.37 1.77 1.29 1.48 4.12 3.50 4.30 3.97 2.17 1.98 2.24 2.13 Extension 0.99 1.43 1.11 1.18 2.97 2.67 3.03 2.89 1.96 2.02 2.15 2.04 Table C.23: Heavyweight abstract objects time for subject #8 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Sph erical 1.73 1.66 1.60 1.66 4.16 4.23 3.57 3.99 3.75 3.51 2.95 3.40 Tripod 1.33 1.37 1.36 1.35 2.61 3.21 2.63 2.82 2.71 2.57 2.55 2.61 Power 1.33 1.43 1.20 1.32 4.22 4.30 3.69 4.07 2.76 2.50 2.51 2.59 Lateral 1.43 1.43 1.23 1.36 3.21 2.85 2.83 2.96 2.65 2.73 2.60 2.66 Tip 1.29 1.29 1.13 1.24 3.76 5.48 3.26 4.17 4.53 2.46 2.12 3.04 Extension 1.28 1.38 1.61 1.42 3.06 2.92 4.87 3.62 2.41 2.40 2.23 2.35 Appendix C: (Continued)

PAGE 109

97 Table C.24: Activities of daily living time data for subject #8 Anatomical hand (sec) Hook (sec ) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 5.45 6.18 4.99 5.54 16.64 20.69 12.85 16.73 19.87 14.93 23.13 19.31 Undo buttons 9.12 4.92 6.46 6.83 23.36 27.12 16.1 8 22.22 49.00 52.61 35.22 45.61 Food cutting 3.17 2.52 2.26 2.65 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Page turning 1.58 1.50 1.32 1.47 3.92 3.09 2.73 3.25 4.51 3.76 3.40 3.89 Jar lid 1.58 1.39 1.49 1.49 3.37 4.29 2.80 3.49 5.63 5.27 3.62 4.84 Jug po ur 3.47 4.50 3.83 3.93 11.98 13.98 9.67 11.88 9.65 6.58 5.65 7.29 Carton pour 6.05 5.15 4.55 5.25 12.43 17.24 12.77 14.15 15.29 15.84 11.86 14.33 Full jar 1.67 1.51 1.42 1.53 5.83 6.38 5.56 5.92 3.46 3.95 3.29 3.57 Empty tin 1.27 1.26 1.24 1.26 2.86 2. 57 2.28 2.57 2.55 2.69 2.27 2.50 Tray 3.14 2.49 2.66 2.76 10.38 10.05 7.04 9.16 8.28 8.93 6.51 7.91 Turn a key 1.10 1.03 1.01 1.05 3.02 2.97 2.62 2.87 2.30 2.72 2.46 2.49 Zipper 1.60 1.62 1.50 1.57 4.59 3.91 4.21 4.24 24.47 11.94 12.07 16.16 Screw 2.8 3 2.89 2.48 2.73 10.12 8.28 8.68 9.03 15.53 12.65 8.21 12.13 Door handle 0.98 0.82 0.87 0.89 1.37 1.30 1.19 1.29 1.87 1.69 1.56 1.71 Appendix C: (Continued)

PAGE 110

98 Table C .25: Lightweight abstract o bjects time for s ubject #9 A natomical hand (sec) H ook (sec) H ook with finge rtips (sec) Task Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Spherical 1.48 1.51 1.33 1.44 6.53 5.89 7.47 6.63 5.01 4.91 5.02 4.98 Tripod 1.94 1.84 1.79 1.86 4.33 5.93 5.91 5.39 4.64 5.47 5.00 5.04 Po wer 1.62 1.81 1.73 1.72 5.29 4.84 5.06 5.06 4.19 5.04 4.90 4.71 Lateral 1.96 1.93 2.07 1.99 4.30 5.05 4.39 4.58 3.97 4.92 4.09 4.33 Tip 1.91 2.11 2.13 2.05 6.73 4.69 4.03 5.15 5.66 5.41 5.07 5.38 Extension 2.54 2.48 2.57 2.53 5.27 4.89 6.12 5.43 5.18 5. 43 6.75 5.79 Table C .26: Heavyweight abstract o bjects time for s ubject #9 A natomical hand (sec) H ook (sec) H ook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Spherical 2.19 2.40 2.25 2.28 0.00 0.00 0.00 0.00 6.02 5.79 6.28 6.03 Tripod 1.98 2.01 1.89 1.96 5.53 5.09 5.59 5.40 5.03 5.07 5.22 5.11 Power 2.26 2.47 2.38 2.37 0.00 0.00 0.00 0.00 4.98 4.69 5.01 4.89 Lateral 2.23 2.60 2.81 2.55 6.44 5.26 6.21 5.97 6.16 6.83 6.54 6 .51 Tip 2.37 2.63 2.20 2.40 6.45 5.13 5.54 5.71 9.28 8.13 8.42 8.61 Extension 3.26 2.60 2.96 2.94 7.82 4.83 5.35 6.00 9.22 7.59 7.05 7.95 Appendix C: (Continued)

PAGE 111

99 Table C .27 : Activities of daily living time data for s ubject #9 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Coins 6.49 5.93 6.43 6.28 25.33 24.95 18.49 22.92 19.25 18.92 24.53 20.90 Undo buttons 7.53 7.11 6.92 7.19 34.24 30.35 31.29 31.96 39. 27 35.96 29.03 34.75 Food cutting 7.43 7.49 7.54 7.49 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Page turning 2.43 2.33 2.28 2.35 7.83 7.98 7.07 7.63 6.41 5.92 5.22 5.85 Jar lid 2.83 2.89 2.82 2.85 17.43 11.80 10.63 13.29 8.33 6.99 5.69 7.00 Jug pour 6.59 5.84 5.63 6.02 13.44 10.90 10.19 11.51 12.39 11.24 10.19 11.27 Carton pour 8.74 9.24 9.36 9.11 18.72 19.65 18.82 19.06 21.56 21.38 20.31 21.08 Full jar 2.74 2.82 2.95 2.84 6.27 6.31 5.98 6.19 5.88 6.32 6.24 6.15 Empty tin 2.48 2.62 2.53 2.54 4.86 4.55 4.89 4.77 5.19 5.51 4.55 5.08 Tray 5.21 4.83 4.51 4.85 10.23 9.89 12.35 10.82 18.42 12.25 16.92 15.86 Turn a key 1.79 1.97 1.65 1.80 3.14 4.09 3.35 3.53 4.50 3.85 3.87 4.07 Zipper 3.55 3.87 3.14 3.52 11.68 9.67 8.59 9.98 15.61 9.58 11.67 12.29 Screw 5.57 6.02 5.22 5.60 10.39 11.49 10.41 10.76 11.65 10.69 11.17 11.17 Door handle 2.00 2.05 1.71 1.92 3.24 3.22 2.67 3.04 3.86 4.03 3.92 3.94 Appendix C: (Continued)

PAGE 112

100 Table C .28: Lightweight a bstract o bjects time for s ubject #10 Anatomical hand (sec) Hook (sec) Hook wi th fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Spherical 1.47 1.45 1.59 1.50 7.42 6.54 6.42 6.79 6.37 5.31 4.76 5.48 Tripod 1.84 1.61 1.63 1.69 9.28 8.23 7.35 8.29 7.31 5.37 5.21 5.96 Power 2.07 1.93 2.05 2.02 9.42 7.69 8.58 8.56 5.39 5.04 5.17 5.20 Lateral 2.27 2.24 2.27 2.26 10.85 8.69 8.97 9.50 7.03 8.03 5.91 6.99 Tip 2.23 2.34 1.99 2.19 8.72 7.11 7.32 7.72 7.93 10.17 7.96 8.69 Extension 2.76 2.97 2.84 2.86 6.87 7.28 7.04 7. 06 6.67 6.17 6.23 6.36 Table C .29 : Heavyweight a bstract o bjects time for s ubject #10 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Trial 1 Trial 2 Trial 3 Average Sp herical 1.87 1.85 2.05 1.92 9.70 9.12 15.00 11.27 6.37 7.38 5.47 6.41 Tripod 1.71 1.59 1.87 1.72 7.75 7.14 7.76 7.55 5.47 5.48 4.39 5.11 Power 2.01 1.95 2.05 2.00 11.23 9.86 8.49 9.86 6.83 5.18 5.21 5.74 Lateral 2.62 2.67 2.53 2.61 8.67 8.01 6.99 7.89 0 .00 0.00 0.00 0.00 Tip 2.03 2.19 2.17 2.13 9.19 6.84 7.59 7.87 7.37 8.63 8.23 8.08 Extension 2.63 3.19 2.65 2.82 7.77 7.36 7.07 7.40 12.87 11.53 8.33 10.91 Appendix C: (Continued)

PAGE 113

101 Table C .30 : Activity of daily living time data for s ubject #10 Anatomical hand (sec) Hook (sec) Hook with fingertips (sec) Task Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Trial 1 Trial 2 Trial 3 A verage Coins 7.12 6.37 6.03 6.51 30.43 29.00 25.78 28.40 33.43 28.78 29.82 30.68 Undo buttons 10.49 9.03 8.57 9.36 41.83 23.0 7 24.68 29.86 0.00 0.00 0.00 0.00 Food cutting 7.90 7.61 6.05 7.19 32.59 52.34 0.00 28.31 0.00 0.00 0.00 0.00 Page turning 2.62 2.67 2.44 2.58 9.29 9.50 7.41 8.73 9.36 9.85 7.93 9.05 Jar lid 3.28 3.21 3.29 3.26 0.00 0.00 0.00 0.00 8.49 7.94 7.49 7.97 J ug pour 6.99 6.29 5.72 6.33 24.67 18.49 15.66 19.61 15.57 26.27 17.18 19.67 Carton pour 10.37 9.49 9.28 9.71 28.56 31.21 25.68 28.48 24.75 28.63 22.63 25.34 Full jar 3.20 3.06 3.07 3.11 12.66 6.89 8.49 9.35 10.34 8.63 7.83 8.93 Empty tin 2.74 2.72 2.57 2.68 7.02 6.63 7.26 6.97 5.90 5.67 5.33 5.63 Tray 4.87 5.22 4.32 4.80 15.93 12.53 11.14 13.20 21.13 14.95 11.26 15.78 Turn a key 2.88 2.63 2.59 2.70 6.83 5.71 5.05 5.86 5.61 4.09 3.37 4.36 Zipper 4.66 4.26 4.89 4.60 19.21 12.53 8.71 13.48 0.00 0.00 0. 00 0.00 Screw 6.35 5.53 4.97 5.62 17.86 12.29 14.23 14.79 13.05 13.12 13.86 13.34 Door handle 3.13 3.19 3.03 3.12 6.02 5.62 5.24 5.63 5.00 5.43 4.29 4.91 Appendix C: (Continued)

PAGE 114

102 Figure D.1: Mean values of the spherical prehensile pattern for each subject Appendix D: Mean valu es of each subject separated by prehensile pattern

PAGE 115

103 Figure D. 2: Mean values of the tripod prehensile pattern for each subject Appendix D : (Continued)

PAGE 116

104 Figure D.3: Mean values of the power prehensile pattern for each subject Appendix D : (Continued)

PAGE 117

105 Figure D.4: Mean values of the lateral prehensile pattern for each subject Appendix D : (Continued)

PAGE 118

106 Figure D.5: Mean values of the tip prehensile pattern for each subject Appendix D : (Continued)

PAGE 119

107 Figure D.6: Mean values of the extension prehensile pattern for each subject Appendix D : (Continued)


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001935133
003 fts
005 20080421132545.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 080421s2007 flua sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002273
040
FHM
c FHM
035
(OCoLC)225864686
049
FHMM
090
TJ145 (ONLINE)
1 100
Ramirez, Issa A.
0 245
Linkage-based prosthetic fingertips :
b analysis and testing
h [electronic resource] /
by Issa A. Ramirez.
260
[Tampa, Fla.] :
University of South Florida,
2007.
3 520
ABSTRACT: This thesis consists of the research on linkage-based fingertips for prosthetic hands. These fingertips consists of small polycentric mechanisms attached to what would be the pulp in normal anatomical fingers. These mechanisms allow the prosthetic hand to conform to the shape of objects during grasp. The goal of these prosthetic fingertips is to maximize the functionality of the hand while minimizing the number of inputs that the user has to control. The stability of the fingertip mechanisms is analyzed using the principle of virtual work. From this analysis we are able to show that the fingertip mechanism is stable for a large range of rotation of the link and for a large range of directions on which the force is applied, and that the mechanism is indifferent to the magnitude of the force applied to it (assuming that the force does not damage/deform the mechanism). To assess if the four-bar mechanisms (fingertips) improve the grasping capabilities in robotics and prosthetics, tests were performed on prosthetic hands and robot grippers with and without the fingertips. Comparisons were made using the Southampton Hand Assessment Procedure (SHAP) protocol, which tests the differences and measures the functionality of particular types of grasp, such as power, spherical, lateral, tripod, tip and extension. In the human testing, the overall Index of Functionality (IOF) of the Hosmer hook is 66.65 and 66.21 for the hook with the fingertips. The hook with the fingertips had a better IOF in the spherical and power prehensile pattern. When the IOF is calculated for the tasks that the fingertips were used, in 10 of 11 of the tasks, the IOF is higher than using the Hosmer hook. In the robotic gripper testing, the Index of Functionality was not be calculated because the time to perform the tasks depended more on the robotic control system than on the physical characteristics of the gripper.
502
Thesis (M.S.)--University of South Florida, 2007.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 107 pages.
590
Advisor: Craig P. Lusk, Ph.D.
653
Linkage-based fingertips.
Stability analysis.
SHAP test.
Gripper.
Hook.
690
Dissertations, Academic
z USF
x Mechanical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2273