Humanoid robot pulls heavy loads with stable whole-body control
Humanoid Rickshaw Pulling: Whole-Body Locomotion under Coupled Wheeled Loads
RoboticsMachine Learning
Summary
Carrying heavy objects can be hard for humanoid robots because they have to balance and hold the load. This paper shows a way for robots to pull wheeled loads, like a rickshaw, which lets them move heavier things more safely. The researchers created a control method that helps the robot keep its balance and hold onto the load even when the forces change unexpectedly. They trained the robot in simulation and then tested it on a real robot, showing it could pull heavy loads up to 115 kg and handle tasks like starting, turning, and stopping without needing to be retrained for each load.
What this means in practice
- •For robotics engineers: Develop control systems for humanoid robots to pull heavy wheeled loads while maintaining balance and stable contact.
- •For logistics automation teams: Automate heavy payload transport using humanoid robots pulling rickshaw-like vehicles in warehouse or urban settings.$Commercial implications: Enables deployable heavy-load carrying humanoid robots to replace or augment human labor in logistics.
Authors
Yangzhi Yang, Xiansheng Lin, Zhaoming Xie, Xiaobin Xiong
Abstract
Humanoid robots could transport payloads substantially heavier than themselves by pulling passive wheeled vehicles instead of carrying the load. This capability, however, creates a coupled locomotion problem: the robot must maintain persistent upper-body contact while adapting to unknown, configuration-dependent forces arising from the payload, vehicle, and terrain. We present a whole-body control framework for humanoid rickshaw pulling that tracks commanded vehicle motion while preserving balance and stable grasps under uncertain load dynamics. During training, a privileged teacher exploits vehicle states, interaction forces, and load properties. Its actions and latent are distilled into a history-conditioned student that implicitly infers coupled dynamics from proprioceptive responses, followed by reinforcement-learning fine-tuning. Comparisons with \emph{No History} and \emph{Only History} baselines show that the resulting policy achieves accurate vehicle tracking while reducing vehicle oscillation, torso tilt, and actuation cost. Behavioral analysis shows that Unitree G1 propels the rickshaw and generates gait-synchronized whole-body reactions that stabilize its lateral and roll motions. Moreover, pulling redistributes joint effort and yields a lower robot-normalized cost-of-transport proxy than unloaded walking over most tested load--speed conditions. On hardware, a single policy performs starting, sustained pulling, turning, and stopping with both rigid payloads and human passengers, handling a loaded rickshaw mass of up to 115~kg without load-specific retuning. These results demonstrate robust heavy-load transportation through coordinated and persistent humanoid--vehicle interaction.