Learning Predictive Models for Ergonomic Control of Prosthetic Devices

GEOFFEY CLARK, Joseph Campbell, Heni Ben Amor
Proceedings of the 2020 Conference on Robot Learning, PMLR 155:2268-2278, 2021.

Abstract

We present Model-Predictive Interaction Primitives – a robot learning framework for assistive motion in human-machine collaboration tasks which explicitly accounts for biomechanical impact on the human musculoskeletal system. First, we extend Interaction Primitives to enable predictive biomechanics: the prediction of future biomechanical states of a human partner conditioned on current observations and intended robot control signals. In turn, we leverage this capability within a model-predictive control strategy to identify the future ergonomic and biomechanical ramifications of potential robot actions. Optimal control trajectories are selected so as to minimize future physical impact on the human musculoskeletal system. We empirically demonstrate that our approach minimizes knee or muscle forces via generated control actions selected according to biomechanical cost functions. Experiments are performed in synthetic and real-world experiments involving powered prosthetic devices.

Cite this Paper

BibTeX
@InProceedings{pmlr-v155-clark21a, title = {Learning Predictive Models for Ergonomic Control of Prosthetic Devices}, author = {CLARK, GEOFFEY and Campbell, Joseph and Amor, Heni Ben}, booktitle = {Proceedings of the 2020 Conference on Robot Learning}, pages = {2268--2278}, year = {2021}, editor = {Kober, Jens and Ramos, Fabio and Tomlin, Claire}, volume = {155}, series = {Proceedings of Machine Learning Research}, month = {16--18 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v155/clark21a/clark21a.pdf}, url = {https://proceedings.mlr.press/v155/clark21a.html}, abstract = {We present Model-Predictive Interaction Primitives – a robot learning framework for assistive motion in human-machine collaboration tasks which explicitly accounts for biomechanical impact on the human musculoskeletal system. First, we extend Interaction Primitives to enable predictive biomechanics: the prediction of future biomechanical states of a human partner conditioned on current observations and intended robot control signals. In turn, we leverage this capability within a model-predictive control strategy to identify the future ergonomic and biomechanical ramifications of potential robot actions. Optimal control trajectories are selected so as to minimize future physical impact on the human musculoskeletal system. We empirically demonstrate that our approach minimizes knee or muscle forces via generated control actions selected according to biomechanical cost functions. Experiments are performed in synthetic and real-world experiments involving powered prosthetic devices.} }
Endnote
%0 Conference Paper %T Learning Predictive Models for Ergonomic Control of Prosthetic Devices %A GEOFFEY CLARK %A Joseph Campbell %A Heni Ben Amor %B Proceedings of the 2020 Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2021 %E Jens Kober %E Fabio Ramos %E Claire Tomlin %F pmlr-v155-clark21a %I PMLR %P 2268--2278 %U https://proceedings.mlr.press/v155/clark21a.html %V 155 %X We present Model-Predictive Interaction Primitives – a robot learning framework for assistive motion in human-machine collaboration tasks which explicitly accounts for biomechanical impact on the human musculoskeletal system. First, we extend Interaction Primitives to enable predictive biomechanics: the prediction of future biomechanical states of a human partner conditioned on current observations and intended robot control signals. In turn, we leverage this capability within a model-predictive control strategy to identify the future ergonomic and biomechanical ramifications of potential robot actions. Optimal control trajectories are selected so as to minimize future physical impact on the human musculoskeletal system. We empirically demonstrate that our approach minimizes knee or muscle forces via generated control actions selected according to biomechanical cost functions. Experiments are performed in synthetic and real-world experiments involving powered prosthetic devices.
APA
CLARK, G., Campbell, J. & Amor, H.B.. (2021). Learning Predictive Models for Ergonomic Control of Prosthetic Devices. Proceedings of the 2020 Conference on Robot Learning, in Proceedings of Machine Learning Research 155:2268-2278 Available from https://proceedings.mlr.press/v155/clark21a.html.

Related Material