Personalized hip exoskeleton assistance adapts to walking speeds efficiently
Learning a Speed-adaptive Hip Exoskeleton Control Policy Via Sim-to-real Reinforcement Learning
Robotics
Summary
Helping people walk better with hip exoskeletons is hard because everyone moves differently and at different speeds. The authors created a new way where a computer learns the best timing for support using simulations. Then, real users help fine-tune how strong the help feels by comparing different options. This approach means fewer trials are needed to find the perfect fit for each person’s walking style.
What this means in practice
- •For rehabilitation engineers: Develop hip exoskeletons that adjust assistance timing and strength personalized to each patient’s walking speed with fewer tuning trials.
- •For robotic exoskeleton developers: Integrate simulation-trained timing policies with real-time user feedback to create adaptable exoskeleton controllers for daily walking assistance.
Authors
Bin Li, Zhimin Hou, Jiacheng Hou, Zenian Liang, Tong Wu, Teng Ma, Chenglong Fu
Abstract
Providing personalized exoskeleton assistance across varying walking speeds remains challenging. Existing online optimization methods are sample-inefficient, requiring extensive human-in-the-loop (HIL) evaluations to optimize the entire assistive torque profile. Sim-to-real reinforcement learning (RL) offers a promising alternative but cannot directly account for individual user preferences. We propose a framework integrating sim-to-real RL with online preference learning for personalized exoskeleton assistance. Specifically, assistance timing is learned in simulation by training RL policies with human musculoskeletal models across varying walking speeds. The learned policies are then distilled and deployed on a physical hip exoskeleton using onboard sensory observations. Gaussian-process-based preference learning further personalizes the assistance magnitude through pairwise user comparisons. By decoupling assistance timing learning in simulation from magnitude optimization in real-world experiments, our framework substantially reduces the online optimization space. Human-subject experiments demonstrate efficient identification of personalized assistive torque profiles across varying walking speeds with fewer real-world evaluations.