Reinforcement Learning for the Soccer Dribbling Task
This work addresses a specific robotic soccer scenario with incremental improvements in dribbling performance.
The paper tackled the soccer dribbling task by proposing a reinforcement learning solution using Sarsa with CMAC, achieving a success rate of 58% against a strong adversary after training.
We propose a reinforcement learning solution to the \emph{soccer dribbling task}, a scenario in which a soccer agent has to go from the beginning to the end of a region keeping possession of the ball, as an adversary attempts to gain possession. While the adversary uses a stationary policy, the dribbler learns the best action to take at each decision point. After defining meaningful variables to represent the state space, and high-level macro-actions to incorporate domain knowledge, we describe our application of the reinforcement learning algorithm \emph{Sarsa} with CMAC for function approximation. Our experiments show that, after the training period, the dribbler is able to accomplish its task against a strong adversary around 58% of the time.