Rapid Human-in-the-Loop Optimization for Hip Exoskeleton Assistance During Walking: An EMG-based Approach

In this study, we evaluated the feasibility of employing electromyography (EMG) as an alternative feedback signal to metabolic cost within a HIL optimization framework. Our findings reveal that, while both approaches yield comparable reductions in net metabolic cost during walking, the EMG--based approach achieves these benefits in only one--quarter of the time required by conventional metabolic cost feedback. By mapping the relationship between EMG and metabolic cost, our results suggest that EMG feedback not only expedites the optimization process but also opens the door to incorporating additional control parameters in the assistance pattern, thereby enhancing overall system complexity and adaptability.