Quadrupedal Robot Skateboard Mounting via Reverse Curriculum Learning

TL;DR

This paper presents a novel reverse curriculum reinforcement learning approach enabling quadrupedal robots to autonomously mount skateboards, starting from simplified conditions and gradually handling more complex, realistic scenarios.

Contribution

The work introduces a reverse curriculum learning method for quadruped skateboard mounting, demonstrating successful transfer from fixed to mobile skateboards with robustness to variations.

Findings

Robust policy learned for skateboard mounting

Successful transfer to mobile skateboard scenarios

Code and models publicly available

Abstract

The aim of this work is to enable quadrupedal robots to mount skateboards using Reverse Curriculum Reinforcement Learning. Although prior work has demonstrated skateboarding for quadrupeds that are already positioned on the board, the initial mounting phase still poses a significant challenge. A goal-oriented methodology was adopted, beginning with the terminal phases of the task and progressively increasing the complexity of the problem definition to approximate the desired objective. The learning process was initiated with the skateboard rigidly fixed within the global coordinate frame and the robot positioned directly above it. Through gradual relaxation of these initial conditions, the learned policy demonstrated robustness to variations in skateboard position and orientation, ultimately exhibiting a successful transfer to scenarios involving a mobile skateboard. The code, trained models, and reproducible examples are available at the following link: https://github.com/dancher00/quadruped-skateboard-mounting