Problem
Research questions and friction points this paper is trying to address.
powered prosthesis
end-to-end control
impedance control
mode classification
subject-specific tuning
End-to-End Control of a Powered Knee-Ankle Prosthesis Towards Unified, Tuning-Free Assistance
🤖 AI Summary
This work addresses the limitations of current powered prostheses, which rely on manually tuned impedance control and explicit gait-mode classification, hindering unified and adaptive, calibration-free assistance. The authors propose an end-to-end prosthesis controller based on a temporal convolutional network (TCN) that directly maps onboard sensor signals to continuous actuator outputs, eliminating the need for explicit intent recognition or subject-specific parameter tuning. Trained on multi-terrain data from 18 transfemoral amputees and deployed in real time on an embedded system, the controller achieves, for the first time, seamless gait adaptation across level ground, ramps, and stairs, and generalizes to unseen limb-guidance sequences. Experiments demonstrate that it accurately reproduces key biomechanical patterns—such as speed-scaled ankle torque and slope-dependent knee pre-flexion—validating its effectiveness and strong generalization capability.
📝 Abstract
Powered prostheses conventionally rely on impedance controllers that require extensive manual tuning and explicit mode classification. In this work, we present real-time deployment of an end-to-end prosthesis controller that estimates continuous actuator signals from onboard sensors, eliminating the need for intent classifiers and subject-specific tuning. Temporal Convolutional Networks were trained on a multi-terrain dataset from 18 individuals with transfemoral amputation and deployed in real time across five locomotion modes. Four participants (three able-bodied, one with transfemoral amputation) ambulated across level ground, ramp ascent and descent, and stair ascent and descent. During level walking, the deployed controller reproduced the training-data scaling of peak ankle torque with walking speed (deployed 0.85 Nm/kg per m/s, p = 0.001; training 0.96 Nm/kg per m/s, 95% CI [0.42, 1.50], p = 0.002), after excluding one outlier traced to atypical prosthesis loading. During ramp ascent, the controller scaled knee pre-flexion with grade (deployed 2.92 deg/deg, p = 0.027; training 3.30 deg/deg, 95% CI [1.83, 4.77], p < 0.001). During ramp descent, the controller increased resistive knee torque relative to level walking (deployed +0.16 Nm/kg, p < 0.001; training +0.16 Nm/kg, p = 0.008). Seamless stair transitions were generated for both intact- and prosthetic-side-leading sequences in ascent and descent, despite the training data containing only one limb-leading sequence. These results provide initial evidence towards end-to-end control that can provide unified, mode-adaptive prosthetic assistance without subject-specific tuning.
Innovation
Methods, ideas, or system contributions that make the work stand out.
end-to-end control
powered prosthesis
Temporal Convolutional Networks
tuning-free
mode-adaptive assistance
🔎 Similar Papers
No similar papers found.
