DribbleBot: Dynamic Legged Manipulation in the Wild

TL;DR

DribbleBot is a legged robot system capable of dribbling a soccer ball in real-world conditions by using simulation-trained policies transferred to the physical robot, addressing challenges in perception and dynamic control.

Contribution

This work introduces a novel approach for dynamic ball manipulation with legged robots using reinforcement learning and real-world transfer, handling variable terrains and perception constraints.

Findings

Successful real-world ball dribbling in diverse terrains

Effective perception of the ball with onboard cameras

Quadruped robots are suitable for complex manipulation tasks

Abstract

DribbleBot (Dexterous Ball Manipulation with a Legged Robot) is a legged robotic system that can dribble a soccer ball under the same real-world conditions as humans (i.e., in-the-wild). We adopt the paradigm of training policies in simulation using reinforcement learning and transferring them into the real world. We overcome critical challenges of accounting for variable ball motion dynamics on different terrains and perceiving the ball using body-mounted cameras under the constraints of onboard computing. Our results provide evidence that current quadruped platforms are well-suited for studying dynamic whole-body control problems involving simultaneous locomotion and manipulation directly from sensory observations.