Human-in-the-loop Optimisation in Robot-assisted Gait Training

2025-10-07

AI Summary

Individual gait patterns during lower-limb exoskeleton-based gait training exhibit high inter- and intra-subject heterogeneity, hindering effective personalized assistance. Method: We propose a human-in-the-loop online optimization framework leveraging Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to adaptively tune assistive stiffness parameters in real time. The method integrates wearable sensing, dynamic gait monitoring, and model-free controller adaptation—marking the first application of CMA-ES to human–robot co-adaptive rehabilitation control. Results: The algorithm reliably converges to subject-specific stiffness configurations, confirming that intra- and inter-individual variability—and human–robot co-adaptation—are critical determinants of personalized control efficacy. However, behavioral variability limited statistically significant improvements in clinical outcomes, revealing a decoupling between “convergence to personalized parameters” and “manifestation of functional benefits.” This finding provides theoretical grounding for future human-centered optimization paradigms in rehabilitation robotics.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Wearable robots offer a promising solution for quantitatively monitoring gait and providing systematic, adaptive assistance to promote patient independence and improve gait. However, due to significant interpersonal and intrapersonal variability in walking patterns, it is important to design robot controllers that can adapt to the unique characteristics of each individual. This paper investigates the potential of human-in-the-loop optimisation (HILO) to deliver personalised assistance in gait training. The Covariance Matrix Adaptation Evolution Strategy (CMA-ES) was employed to continuously optimise an assist-as-needed controller of a lower-limb exoskeleton. Six healthy individuals participated over a two-day experiment. Our results suggest that while the CMA-ES appears to converge to a unique set of stiffnesses for each individual, no measurable impact on the subjects' performance was observed during the validation trials. These findings highlight the impact of human-robot co-adaptation and human behaviour variability, whose effect may be greater than potential benefits of personalising rule-based assistive controllers. Our work contributes to understanding the limitations of current personalisation approaches in exoskeleton-assisted gait rehabilitation and identifies key challenges for effective implementation of human-in-the-loop optimisation in this domain.

Problem

• Personalizing robot assistance for individual gait training
• Adapting controllers to interpersonal walking pattern variability
• Evaluating human-robot co-adaptation effects on performance outcomes

Innovation

• Human-in-the-loop optimization for gait training
• CMA-ES algorithm adapts exoskeleton controller parameters
• Personalized stiffness settings via continuous optimization process