ANALYTIC UNDERPINNING OF BIOMIMETIC PROSTHETICS


simulate not only the replaced part but the ways in which that part fi ts into the continuum of organisation and environment. Making a prosthesis which minimises the user’s sense of ‘separateness’ requires close analytic attention to the force systems which it generates and to which it responds. Those include the reaction of the ground to a foot placed upon it, the consequent responses of every part of the body whose centre of mass is balanced upon that foot, and conversion of fl exure energy into enhanced motion. Also crucial is understanding of the continually shifting relationship within the organism between centralised and decentralised musculoskeletal control systems, and equally variant coordination between the various decentralised subsystems themselves. The more closely a prosthesis can fi t into those relationships, the more useful it will be. One example[5]


of research in this area that


makes very explicit the inseparability of robotics, prosthesis and biomimesis is Insects running on elastic surfaces, led by Andrew Spence, currently at the Royal Veterinary College’s Structure and Motion Laboratory. Here, connections and common factors are studied across bipedal and hexapodal systems, biological and robotic, in motion on surfaces with a range of compliance in relation to body stiffness. Spence and his colleagues gathered their data using an accelerometer carried as a backpack by cockroaches, placed as close as possible to the insect’s centre of mass (CoM), linked to video tracking, analysing the results in MatLab (using purpose written scripts, the Statistical toolbox, RMAOV33 three-way ANOVA routine and Kalman fi lter) and SPSS. The insects moved across a stiff surface interrupted by a latex membrane, whose stiffness could be varied by stretching under load. The hypothesis emerging from their analysis suggests a feed forward control model with localised perturbations largely met by local mechanical feedback at the limb system level. Interestingly, the same model appears to hold for rigid obstacles (for example, negotiating a step) and energy absorbing surfaces (such as mud or sand). Another conceptually-related study, by Alena and others, investigates ground force


Grabowski[6]


reactions contributing to the spring effect in both biological and running-specifi c artifi cial limbs during athletics – an important factor in the controversy around the Pistorius case with which I opened. Their data analysis (repeated measures ANOVA and Tukey range test) relates laboratory treadmill experiment to high-defi nition video records of top Olympic and Paralympic performances. Their conclusions are surprising in some details, and suggest overall that current prostheses, even at the leading edge of development, impose a limit on the spring effects greater than that experienced by natural limbs. A subtler aspect of biomimetic integration,


not yet addressed by prostheses in production, is the sense of touch which gives feedback to the


data intensive


Simulation of cockroach traversing scaled elastic surface perturbation. Vertical acceleration of the COM (A), and equal time interval snapshots of the model’s progression over a simulated elastic surface (B). (From Spence et al[5]


)


organism about far fi ner and more complex forces than those which govern balance and speed. A very light but unexpected touch to the sole of the foot, for example, may trigger locally autonomous reactions within the leg to relocate – the decision being different for a sharp object or a feather. Use of the hand for manipulation depends upon extremely small and brief stimulation of cutaneous tactile nerves by objects to be manipulated before commitment has been made to a grasp. At a cruder level, force feedback within the


structure of the fi ngers, hand and wrist allows reactive control of the same manipulation once the hand is engaged with the manipulated object. Research is increasingly addressing this, analysing the degree to which different substitute mechanisms can be utilised by the nervous system. Analysis of data from experiments (for example, Marasco[4]


et


al) with amputees who have undergone surgical re-innervation show that pseudocutaneous feedback from pressure and temperature sensors on a haptic prosthesis produces signifi cant results. Both subjective and objective measures of embodiment into self image are increased, embodiment of the prosthesis into the user’s self image enhanced, and utility of the prosthesis brought closer to that of an integrated body component. There is a long way yet to go before anything


in this fi eld approaches perfection, but progress is rapid. Computer modelling of biomimetic prosthetic questions is improving all the time and experimental data analysis is an essential part of the development cycle, providing the raw material for each round of model improvement and testing, validating and revising the results.


Honda’s single leg partial exoskeleton, Walk Assist (Image courtesy of Honda UK)


References and Sources For a full list of references and sources cited in this article, please visit www.scientifi c-computing.com/ features/references/august11.php


Statistics special 19


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48