Learning Predictive Models for Ergonomic Control of Prosthetic Devices
This work addresses the challenge of improving ergonomics and reducing physical strain for prosthetic device users, representing an incremental advance in human-machine collaboration.
The authors tackled the problem of controlling prosthetic devices to minimize biomechanical impact on the human musculoskeletal system by developing a robot learning framework that predicts future biomechanical states and selects optimal control actions. They demonstrated that their approach reduces knee or muscle forces in synthetic and real-world experiments with powered prosthetic devices.
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.