Guiding Energy-Efficient Locomotion through Impact Mitigation Rewards
Authors:
- Chenghao WangCWChenghao Wang
- Arjun ViswanathanAVArjun Viswanathan
- Eric SihiteESEric Sihite
- Alireza RamezaniARAlireza Ramezani
Preprint
Abstract
Animals achieve energy-efficient locomotion by their implicit passive dynamics, a marvel that has captivated roboticists for decades.Recently, methods incorporated Adversarial Motion Prior (AMP) and Reinforcement learning (RL) shows promising progress to replicate Animals' naturalistic motion. However, such imitation learning approaches predominantly capture explicit kinematic patterns, so-called gaits, while overlooking the implicit passive dynamics. This work bridges this gap by incorporating a reward term guided by Impact Mitigation Factor (IMF), a physics-informed metric that quantifies a robot's ability to passively mitigate impacts. By integrating IMF with AMP, our approach enables RL policies to learn both explicit motion trajectories from animal reference motion and the implicit passive dynamic. We demonstrate energy efficiency improvements of up to 32%, as measured by the Cost of Transport (CoT), across both AMP and handcrafted reward structure.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- A Learning Framework for Diverse Legged Robot Locomotion Using Barrier-Based Style Rewards2025
- Quadrupedal Locomotion Control On Inclined Surfaces Using Collocation Method2024
- Multi-Modal Mobility Morphobot (M4) with appendage repurposing for locomotion plasticity enhancementNature Communications, 2023
- Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions2022
- Control of Thruster-Assisted, Bipedal Legged Locomotion of the Harpy RobotFrontiers in Robotics and AI, 2021
- Learning Agile Robotic Locomotion Skills by Imitating Animals2020
- Mode-adaptive neural networks for quadruped motion controlACM Transactions on Graphics, 2018
- Proprioceptive Actuator Design in the MIT Cheetah: Impact Mitigation and High-Bandwidth Physical Interaction for Dynamic Legged RobotsIEEE Transactions on Robotics, 2017
- A Three-Dimensional Passive-Dynamic Walking Robot with Two Legs and KneesThe International Journal of Robotics Research, 2001