Partial Motion Imitation for Learning Cart Pushing with Legged Manipulators
2026-03-27 • Robotics
Robotics
AI summaryⓘ
The authors developed a way to help legged robots move and manipulate objects at the same time, which is usually hard because the robot must stay balanced while doing precise actions. They first trained the robot to walk well under many different conditions. Then, they taught the robot to push a cart by copying just the lower-body movements from the walking skill, using a special imitation method. Their tests showed the robot could push carts accurately on various paths and worked well across different simulation platforms. Compared to other methods, their approach made the robot more stable and precise when moving and manipulating simultaneously.
loco-manipulationlegged robotsimitation learningadversarial motion priordomain randomizationterrain randomizationlocomotion policyMuJoCoIsaacLabmobile manipulation
Authors
Mili Das, Morgan Byrd, Donghoon Baek, Sehoon Ha
Abstract
Loco-manipulation is a key capability for legged robots to perform practical mobile manipulation tasks, such as transporting and pushing objects, in real-world environments. However, learning robust loco-manipulation skills remains challenging due to the difficulty of maintaining stable locomotion while simultaneously performing precise manipulation behaviors. This work proposes a partial imitation learning approach that transfers the locomotion style learned from a locomotion task to cart loco-manipulation. A robust locomotion policy is first trained with extensive domain and terrain randomization, and a loco-manipulation policy is then learned by imitating only lower-body motions using a partial adversarial motion prior. We conduct experiments demonstrating that the learned policy successfully pushes a cart along diverse trajectories in IsaacLab and transfers effectively to MuJoCo. We also compare our method to several baselines and show that the proposed approach achieves more stable and accurate loco-manipulation behaviors.