An Adaptable Approach to Learn Realistic Legged Locomotion without Examples

TL;DR

This paper introduces a flexible reinforcement learning-based method guided by a simple model to generate realistic, energy-efficient legged locomotion across different robot morphologies without relying on motion capture or strict constraints.

Contribution

It presents a generic, adaptable approach that uses a spring-loaded inverted pendulum model to guide RL in learning natural locomotion for various robot sizes and types.

Findings

Successfully generated realistic gaits for bipedal and quadrupedal robots

Achieved energy-efficient locomotion without motion capture or strict kinematic constraints

Applicable to different robot morphologies and control architectures

Abstract

Learning controllers that reproduce legged locomotion in nature has been a long-time goal in robotics and computer graphics. While yielding promising results, recent approaches are not yet flexible enough to be applicable to legged systems of different morphologies. This is partly because they often rely on precise motion capture references or elaborate learning environments that ensure the naturality of the emergent locomotion gaits but prevent generalization. This work proposes a generic approach for ensuring realism in locomotion by guiding the learning process with the spring-loaded inverted pendulum model as a reference. Leveraging on the exploration capacities of Reinforcement Learning (RL), we learn a control policy that fills in the information gap between the template model and full-body dynamics required to maintain stable and periodic locomotion. The proposed approach can be applied to robots of different sizes and morphologies and adapted to any RL technique and control architecture. We present experimental results showing that even in a model-free setup and with a simple reactive control architecture, the learned policies can generate realistic and energy-efficient locomotion gaits for a bipedal and a quadrupedal robot. And most importantly, this is achieved without using motion capture, strong constraints in the dynamics or kinematics of the robot, nor prescribing limb coordination. We provide supplemental videos for qualitative analysis of the naturality of the learned gaits.