Creating a Dynamic Quadrupedal Robotic Goalkeeper with Reinforcement Learning

TL;DR

This paper introduces a hierarchical reinforcement learning framework enabling quadrupedal robots to perform real-time soccer goalkeeping by combining multiple locomotion skills with high-level planning for fast ball interception.

Contribution

The paper presents a novel hierarchical RL approach with multiple control policies and a high-level planner for dynamic quadrupedal goalkeeping tasks in real-world scenarios.

Findings

Effective real-world interception of fast-moving balls

Hierarchical RL enables diverse locomotion skills

Framework achieves agile goalkeeping maneuvers

Abstract

We present a reinforcement learning (RL) framework that enables quadrupedal robots to perform soccer goalkeeping tasks in the real world. Soccer goalkeeping using quadrupeds is a challenging problem, that combines highly dynamic locomotion with precise and fast non-prehensile object (ball) manipulation. The robot needs to react to and intercept a potentially flying ball using dynamic locomotion maneuvers in a very short amount of time, usually less than one second. In this paper, we propose to address this problem using a hierarchical model-free RL framework. The first component of the framework contains multiple control policies for distinct locomotion skills, which can be used to cover different regions of the goal. Each control policy enables the robot to track random parametric end-effector trajectories while performing one specific locomotion skill, such as jump, dive, and sidestep. These skills are then utilized by the second part of the framework which is a high-level planner to determine a desired skill and end-effector trajectory in order to intercept a ball flying to different regions of the goal. We deploy the proposed framework on a Mini Cheetah quadrupedal robot and demonstrate the effectiveness of our framework for various agile interceptions of a fast-moving ball in the real world.