Efficient Learning-Based Control of a Legged Robot in Lunar Gravity

2025-09-12

AI Summary

To address the low energy efficiency and poor gravity adaptability of legged robots in low-gravity environments (e.g., Moon, Mars, asteroids), this paper proposes a reinforcement learning–based gravity-adaptive control framework. We innovatively design a gravity-scaled power-optimization reward function to enable policy transfer across diverse gravitational fields—from lunar gravity (0.16 g) to hyper-Earth conditions (up to 2 g). By integrating analytical dynamics modeling with end-to-end policy training, we enhance locomotor energy efficiency. A spring-based constant-force unloading system is developed to physically emulate lunar gravity (0.16 g) on Earth for experimental validation. Results show that at 0.4 m/s walking speed under Earth gravity, power consumption is reduced by 23% to 23.4 W versus baseline; under lunar gravity, it further drops to 12.2 W—a 36% reduction. This work is the first to jointly integrate gravity-scaled reward shaping with a physically verifiable unloading platform, significantly improving motion energy efficiency and environmental adaptability of planetary rovers under stringent power and thermal constraints.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Legged robots are promising candidates for exploring challenging areas on low-gravity bodies such as the Moon, Mars, or asteroids, thanks to their advanced mobility on unstructured terrain. However, as planetary robots' power and thermal budgets are highly restricted, these robots need energy-efficient control approaches that easily transfer to multiple gravity environments. In this work, we introduce a reinforcement learning-based control approach for legged robots with gravity-scaled power-optimized reward functions. We use our approach to develop and validate a locomotion controller and a base pose controller in gravity environments from lunar gravity (1.62 m/s2) to a hypothetical super-Earth (19.62 m/s2). Our approach successfully scales across these gravity levels for locomotion and base pose control with the gravity-scaled reward functions. The power-optimized locomotion controller reached a power consumption for locomotion of 23.4 W in Earth gravity on a 15.65 kg robot at 0.4 m/s, a 23 % improvement over the baseline policy. Additionally, we designed a constant-force spring offload system that allowed us to conduct real-world experiments on legged locomotion in lunar gravity. In lunar gravity, the power-optimized control policy reached 12.2 W, 36 % less than a baseline controller which is not optimized for power efficiency. Our method provides a scalable approach to developing power-efficient locomotion controllers for legged robots across multiple gravity levels.

Problem

• Developing energy-efficient legged robot control for lunar gravity exploration
• Creating gravity-scaled reinforcement learning controllers for multiple planetary environments
• Reducing power consumption in legged robots across varying gravity levels

Innovation

• Reinforcement learning-based control approach
• Gravity-scaled power-optimized reward functions
• Constant-force spring offload system