HWC-Loco: A Hierarchical Whole-Body Control Approach to Robust Humanoid Locomotion

Abstract

Humanoid robots, capable of assuming human roles in various workplaces, have become essential to the advancement of embodied intelligence. However, as robots with complex physical structures, learning a control model that can operate robustly across diverse environments remains inherently challenging, particularly under the discrepancies between training and deployment environments. In this study, we propose HWC-Loco, a robust whole-body control algorithm tailored for humanoid locomotion tasks. By reformulating policy learning as a robust optimization problem, HWC-Loco explicitly learns to recover from safety-critical scenarios. While prioritizing safety guarantees, overly conservative behavior can compromise the robot's ability to complete the given tasks. To tackle this challenge, HWC-Loco leverages a hierarchical policy for robust control. This policy can dynamically resolve the trade-off between goal-tracking and safety recovery, guided by human behavior norms and dynamic constraints. To evaluate the performance of HWC-Loco, we conduct extensive comparisons against state-of-the-art humanoid control models, demonstrating HWC-Loco's superior performance across diverse terrains, robot structures, and locomotion tasks under both simulated and real-world environments.

Method

Overview of HWC-Loco: The framework consists of two stages: (a) Training goal-tracking policy to effectively enable human-like locomotion across diverse terrains and safety recovery policy to recover from safety-ciritical states (i.e., extreme-case). (b) Training the high-level planning policy to select between the two pre-trained low-level policies, thereby ensuring locomotion stability and consistency.

Experiments in Simulation

Experiments in Real World

Conclusion

We introduce HWC-Loco, a hierarchical control framework for humanoid robots that incorporates an embedded safety recovery mechanism. This framework has been validated across various locomotion tasks, demonstrating exceptional scalability, robustness, naturalness, and adaptability across diverse tasks and scenarios. Notably, the safety mechanism in HWC-Loco extends beyond locomotion, enabling reliable performance in complex tasks through a dynamic task-safety balance. This ensures robust operation in real-world deployments, positioning HWC-Loco as a foundational solution for safety-critical applications such as industrial automation and assistive robotics. A promising direction for future research is integrating HWC-Loco with upper-body manipulation skills, enabling safety-critical control across a broader range of tasks involving different objects.

Citation

                @article{lin2025hwc,
                    title={HWC-Loco: A Hierarchical Whole-Body Control Approach to Robust Humanoid Locomotion},
                    author={Lin, Sixu and Qiao, Guanren and Tai, Yunxin and Li, Ang and Jia, Kui and Liu, Guiliang},
                    journal={arXiv preprint arXiv:2503.00923},
                    year={2025}
                  }