Quadruped robots adapt gaits to save energy on tough terrain
Energy-Efficient Gait Adaptation via Hierarchical Reinforcement Learning for Quadrupedal Locomotion Across Diverse Terrains
RoboticsMachine Learning
Summary
Walking robots use a lot of energy, especially when moving over different surfaces or at different speeds. The authors created a two-part teaching system where one part controls quick, steady leg movements and the other changes the walking style to save energy. Their method helps robots automatically switch gaits depending on how fast they are moving, from slow pacing to faster trotting. Tests showed the robots use less energy and walk sturdily on flat, rough, and sloped grounds. This approach also worked well on a real robot without extra tuning.
What this means in practice
- •For robotic system integrators: Create energy-saving walking patterns for four-legged robots operating in varied terrain and speeds.
- •For robotics product developers: Deploy quadruped robots with improved battery life and versatile movement across outdoor and uneven environments.$Commercial implications: This enables robotic companies to offer longer-lasting four-legged robots for industrial inspection or delivery in rough terrain.
Authors
Ammar Issa, Anubhav Singh, Anton Tsaritsin, Sergey Kolyubin
Abstract
While energy efficiency is a critical objective for legged-robot locomotion control, achieving low energy consumption while maintaining robust performance across different velocity ranges and terrain conditions remains a key challenge. This is particularly true for end-to-end RL policies, where gait generation, motion execution, and energy optimization are tightly coupled, leading to high sensitivity to reward design. In this work, we propose a hierarchical reinforcement learning (HRL) framework that separates a high-frequency policy for stable and robust joint-level motion execution from low-frequency gait adaptation that explicitly minimizes the cost of transport (CoT). The three-stage Isaac-based training procedure enables zero-shot sim-to-real transfer with improved tracking accuracy, robustness, and energy efficiency. The learned hierarchy exhibits automatic speed-dependent gait adaptation, transitioning from pacing at low speeds to trotting at higher speeds. We validate the proposed approach in simulation against representative single-policy and hierarchical locomotion baselines, demonstrating reduced CoT over a broad range of commanded velocities, while maintaining robust locomotion across flat, uneven rough, and inclined terrains. We further demonstrate its practical feasibility through zero-shot deployment on a physical Unitree AlienGo quadruped.