High precision robot walking improved with world model guided control
LocoWM: High-Precision Locomotion through World-Model-Guided Residual Adaptation
Robotics
Summary
Moving robots precisely so they can perform tasks and handle unexpected bumps is hard. The authors present a method called LocoWM that predicts what the robot will do next and adjusts its movements ahead of time. This way, the robot can correct its actions before errors happen, instead of reacting afterwards. Their tests show this approach helps robots walk more accurately and recover better from pushes and uneven ground.
What this means in practice
- •For robotics engineers: Develop robot locomotion controllers that adjust movements in advance for improved accuracy on rough terrain and under disturbances.
- •For automation integrators: Implement more precise walking robots for automation tasks requiring reliable movement near delicate objects or uneven surfaces.
Authors
Zijie Zhao, Shengqian Chen, Xiaoxu Wang, Han Jiang, Yuanheng Zhu, Dongbin Zhao
Abstract
High-precision locomotion combines motion-command tracking with precise regulation of task-relevant physical states, enabling robots to interact reliably with their surroundings during motion. Joint end-to-end optimization can leave precision objectives insufficiently optimized, while reactive residual control adjusts actions only after deviations become observable. We present \textbf{LocoWM}, a world-model-guided preactive residual adaptation framework for high-precision locomotion. A base policy provides command-following locomotion, while an action-conditioned world model predicts a sequence of future physical states from proprioceptive history and the proposed base action. A residual adapter conditions on this predicted sequence to generate additive action corrections that compensate for anticipated deviations. Two-stage training first learns locomotion and action-conditioned dynamics, then freezes both modules while training the adapter, separating locomotion acquisition from precision adaptation. Experiments spanning terrain leveling, acceleration compensation, and push recovery demonstrate improved control precision and disturbance robustness over end-to-end and reactive residual baselines. Demos and code are available at: https://zhaozijie2022.github.io/LocoWM