Energy-optimal whole-body locomotion control framework for hydraulic quadruped robots via pump-valve coordination

Hydraulic quadruped robots exhibit exceptional power density and dynamic robustness, while energy efficiency remains a critical bottleneck constraining operational endurance. This paper proposes an energy-optimal whole-body locomotion control framework based on pump-valve coordination to improve energy efficiency and maintain stable locomotion performance for the robot. First, a multi-source energetic model is established to systematically characterize the energy transduction and dissipation mechanisms. Building upon this, an energy-aware predictive planner enforces the minimum pressure required by future motion sequences as a spatiotemporal constraint to concurrently optimize supply pressures and locomotive trajectories, while a hierarchical whole-body controller leverages null-space projection and an adaptive regulator to ensure the synergetic execution of pump-valve commands. Experimental results demonstrate that the proposed framework achieves a 27.23% improvement in energy efficiency over the constant-pressure strategy without compromising stability. Furthermore, the method exhibits robust adaptability, consistently reducing the cost of transport across diverse velocities and payloads. The framework serves as a valuable reference for future research on the energy efficiency of quadruped robots in complex environments.