RealAnt: An Open-Source Low-Cost Quadruped for Education and Research in Real-World Reinforcement Learning

TL;DR

RealAnt is an affordable, easy-to-assemble quadruped robot platform designed for reinforcement learning research and education, capable of learning to walk quickly and supporting simulation and open-source sharing.

Contribution

We introduce RealAnt, a low-cost, easy-to-build quadruped robot with validated RL benchmarks, and provide open-source hardware, software, and simulation models for research and education.

Findings

RealAnt costs approximately 350 EUR and can be assembled in under an hour.

The robot can learn to walk from scratch in less than 10 minutes of experience.

Open-source hardware and simulation models facilitate reproducible research and educational use.

Abstract

Current robot platforms available for research are either very expensive or unable to handle the abuse of exploratory controls in reinforcement learning. We develop RealAnt, a minimal low-cost physical version of the popular `Ant' benchmark used in reinforcement learning. RealAnt costs only $\sim$350 EUR (\$410) in materials and can be assembled in less than an hour. We validate the platform with reinforcement learning experiments and provide baseline results on a set of benchmark tasks. We demonstrate that the RealAnt robot can learn to walk from scratch from less than 10 minutes of experience. We also provide simulator versions of the robot (with the same dimensions, state-action spaces, and delayed noisy observations) in the MuJoCo and PyBullet simulators. We open-source hardware designs, supporting software, and baseline results for educational use and reproducible research.