Exploiting Hierarchical Controller Structure in Contextual Parameter Learning for Humanoid Loco-Manipulation

2026-10-03

AI Summary

This study addresses the limitation of independent parameter tuning in hierarchical control for humanoid robots, which neglects cross-layer interactions. To overcome this, we propose a contextual Bayesian optimization framework for jointly learning controller parameters. Methodologically, by leveraging layer-wise performance metrics, parameter coupling, and varying operating conditions inherent to the hierarchy, we construct a correlated multi-output Gaussian process model and design an analytical aggregated objective function. Experimental evaluations on a box-pushing task under varying mass conditions demonstrate that the proposed approach achieves the lowest average empirical regret among compared methods. These results confirm its effectiveness in significantly enhancing cross-layer collaborative optimization efficiency for hierarchical robot control.

Abstract

Hierarchical control architectures are widely used to decompose complex control problems into interacting control levels and are particularly important in robotics, where planning, whole-body motion, and lower-level control must be coordinated across different levels of abstraction and time scales. Their overall closed-loop performance, however, depends strongly on parameters distributed across the hierarchy, such that tuning controllers on different levels independently may neglect relevant cross-layer interactions. We propose a contextual Bayesian optimization framework for joint parameter learning in hierarchical control systems. Rather than modeling closed-loop performance only as a scalar black-box function, we retain separate observations of task performance, realization quality, and control effort. A correlated multi-output Gaussian process models these performance components, while their known aggregation into the overall closed-loop objective is evaluated analytically. The formulation exploits three complementary consequences of hierarchical control: informative performance quantities exposed by the hierarchy, coupling between parameters of different controller levels, and variations of these relations with operating conditions. We evaluate the approach for humanoid loco-manipulation, jointly tuning a centroidal predictive controller and a whole-body controller for physical box pushing under varying box mass. The proposed method achieves the lowest mean empirical regret during both training and adaptation among the considered baselines.

Problem

• Hierarchical control
• Parameter tuning
• Bayesian optimization
• Humanoid loco-manipulation
• Cross-layer interaction

Innovation

• Contextual Bayesian Optimization
• Hierarchical Control
• Multi-output Gaussian Process
• Humanoid Loco-manipulation
• Joint Parameter Learning