OmniMimic improves quadruped robot walking in all directions
OmniMimic: Dynamics-completed Motion Augmentation for Multi-style Omnidirectional Quadruped Locomotion
RoboticsComputer Vision and Pattern Recognition
Summary
Quadruped robots can learn to walk like animals, but usually only forward, because animal videos show them moving mostly in that direction. The authors introduce OmniMimic, which creates new training data by flipping and reversing these animal motions so robots can also walk backward, sideways, and turn naturally. It combines these new motions with a smart control system that balances general walking skills with specific style adjustments. This helps robots follow commands in many directions more accurately than previous methods.
What this means in practice
- •For quadruped robot developers: Create multi-directional walking controllers that use limited animal motion data to produce diverse, natural gaits in all directions.
- •For robot simulation engineers: Generate augmented motion datasets for testing and validating omnidirectional quadruped locomotion across various gait styles.
Authors
Sheng Wu, Guoqiang Zhao, Zhe Yang, Fei Teng, Zhikun Zhou, Yanlin Yang, Zheng Fang, Hong Zheng, Yaonan Wang, Kailun Yang
Abstract
Animal demonstrations provide quadruped robots with natural and distinctive gait styles that are difficult to specify through hand-crafted rewards. However, their narrow directional coverage leaves little style-consistent supervision for backward, lateral, and turning commands. We present OmniMimic, a training framework that turns directionally limited animal demonstrations into a single multi-gait policy over target per-axis velocity ranges. OmniMimic first combines temporal reversal, constrained dynamics completion, and sagittal reflection to construct robot-specific kinematic and physical supervision beyond the observed directions. It then expands commands progressively from the demonstrated velocity distribution toward the target per-axis bounds, and uses a shared actor with soft-gated, gait-specialized residual experts to balance reusable locomotion skills with gait-specific corrections. Across four gaits in simulation, OmniMimic reduces mean foot-position RMSE at forward and backward reference velocities by 12.9% and velocity-tracking RMSE on a uniform Cartesian command grid by 63.1%, compared with the matched APEX baseline. The project page is at https://OmniMimic.github.io.