# Learning a Speed-adaptive Hip Exoskeleton Control Policy Via Sim-to-real Reinforcement Learning

arXiv: https://arxiv.org/abs/2609.28027 · HTML: https://mlsanity.com/paper/2609.28027/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 8 of its 32 references that are in the map.
- First posted on arXiv 23 Sep 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 8 of 8)

- [HumanMimic: Learning Natural Locomotion and Transitions for Humanoid Robot via Wasserstein Adversarial Imitation](https://mlsanity.com/paper/2309.14225/) (2023)
- [Self Model for Embodied Intelligence: Modeling Full-Body Human Musculoskeletal System and Locomotion Control with Hierarchical Low-Dimensional Representation](https://mlsanity.com/paper/2312.05473/) (2023)
- [Human Preference-Based Learning for High-dimensional Optimization of Exoskeleton Walking Gaits](https://mlsanity.com/paper/2003.06495/) (2020)
- [Learning Speed-Adaptive Walking Agent Using Imitation Learning with Physics-Informed Simulation](https://mlsanity.com/paper/2412.03949/) (2024)
- [Towards Expedited Impedance Tuning of a Robotic Prosthesis for Personalized Gait Assistance by Reinforcement Learning Control](https://mlsanity.com/paper/2006.06518/) (2020)
- [Rapid Online Learning of Hip Exoskeleton Assistance Preferences](https://mlsanity.com/paper/2502.15366/) (2025)
- [Learning Hip Exoskeleton Control Policy via Predictive Neuromusculoskeletal Simulation](https://mlsanity.com/paper/2603.04166/) (2026)
- [Real-Time Gait Phase and Task Estimation for Controlling a Powered Ankle Exoskeleton on Extremely Uneven Terrain](https://mlsanity.com/paper/2205.00155/) (2022)

## Abstract (arXiv)

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.