Hierarchical Control for Robust Standing Stability and Fall Recovery of Task-Performing Humanoid Robots

Maintaining robust standing stability is critical for humanoid robots performing precision manipulation tasks, where traditional balance strategies (e.g., stepping) can disrupt task execution. Designing a controller that ensures both stability and effective disturbance rejection is challenging, as these goals impose conflicting requirements on body sway and stepping behavior. To address this, we propose a hierarchical control framework that decouples these objectives into three parts: 1) a Task-Oriented Standing Policy trained via reinforcement learning with strict penalties on stepping and body sway, specialized for non-stepping disturbance rejection; 2) Ankle Strategy Modulation embedded within the standing policy, which dynamically adjusts the Zero-Moment Point (ZMP) to eliminate Divergent Component of Motion (DCM) errors, thereby accelerating recovery from disturbances; and 3) a fall prediction and recovery mechanism that triggers a transition to a robust stepping policy when disturbances exceed the stability threshold. Extensive experiments on a Q-series humanoid robot, in both simulation and hardware, validate the controller’s effectiveness. We achieved a 100% task success rate over 100 trials in practical scenarios including robot archery and part grasping/placement. Note to Practitioners—This study addresses a critical challenge in deploying humanoid robots for manipulation tasks: maintaining operational continuity during frequent upper-body movements, where traditional balance strategies—such as stepping—can disrupt workflow. We propose a hierarchical control framework that resolves the inherent conflict between non-stepping stability and fall recovery through three integrated components: a reinforcement learning-based standing policy enforcing minimal body sway and zero foot movement, ankle-based Zero-Moment Point (ZMP) modulation for real-time Dynamic Capture Model (DCM) error correction, and a fall prediction mechanism that triggers stepping recovery when necessary. This architecture enables separate training of non-stepping and stepping strategies, improving both stability and disturbance resilience. Experimental validation demonstrates 100% task success over 100 trials across realistic tasks, such as archery and part manipulation. Limitations include simplified sensory, and disturbance models and the decoupled control of upper and lower body motions, which constrain whole-body coordination. Future work will focus on improving multimodal perception and implementing whole-body cooperative control for more complex tasks.