Humanoid robot body control improves stability with separate smoothness limits
Smoothness as a Constraint for Stable Humanoid Locomotion
Robotics
Summary
Controlling walking robots is tricky because different parts of the body need different levels of smooth movement. The legs must react quickly to keep balance, while the upper body should move smoothly to stay steady. The authors created a new method called DeCap that sets separate smoothness rules for the legs and upper body, making the robot walk more stable and smooth. This approach also avoids complicated tuning and works well on different terrains.
What this means in practice
- •For robotics engineers: Implement whole-body control policies that keep humanoid robots stable with separate smoothness controls for upper and lower body parts.
- •For robotic terrain navigation teams: Use fixed smoothness constraints to adapt humanoid robot walking control across different ground surfaces without extra tuning.
Authors
Utsav Panchal, Denis Kleyko, Unal Artan, Amy Loutfi
Abstract
Embodied AI systems, particularly humanoid robots deployed in real world scenarios require whole-body control policies that are both task-responsive and physically smooth. However, smoothness is not uniform across the body: lower body must remain sufficiently reactive, while the upper body must be tightly regulated to preserve stability. Existing reinforcement learning approaches typically impose smoothness through auxiliary terms in the reward function, which compete with task objectives, treating the body as uniform and provide no direct control over the physical quantities responsible for smooth behavior. We introduce DeCap (Decoupled Constraint-aware policy), a constrained reinforcement learning algorithm that decouples whole-body smoothness into separate upper- and lower-body constraint groups, each formulates smoothness as explicit constraints on physical motion limits. To improve constraint satisfaction near feasibility boundaries, DeCap incorporates a bounded barrier penalty that activates proactively as limits are approached and remains bounded at the constraint limit. On real-world humanoid whole-body control task, DeCap reduces upper-body action rate by 2.50x and acceleration by 2.18x relative to reward-based smoothness policies, while also improving lower-body smoothness and reducing transient motion. We demonstrate that a fixed set of smoothness constraints transfers across diverse terrains, alleviating the need of extensive reward tuning.