DribbleBot: 4 Legged Robotic System for Playing Soccer with Reinforcement Learning

In the past few years, we have covered plenty of bipedal robots for playing soccer. The DribbleBot is a 4-legged robotic system from MIT’s Improbable Artificial Intelligence Lab, part of the Computer Science and Artificial Intelligence Laboratory (CSAIL) that can maneuver a soccer ball on sand, gravel, mud, and snow. It uses reinforcement learning to adapt.

This robot can navigate unfamiliar terrains and recover from falls. It carries a bunch of sensors to perceive its surroundings and understand its position. The above video shows it in action.

[HT]