Learning to Drive using Inverse Reinforcement Learning and Deep Q-Networks
This is an incremental improvement for autonomous driving simulation, addressing reward extraction in complex environments.
The paper tackled the problem of extracting reward functions in large state spaces for autonomous driving using inverse reinforcement learning with Deep Q-Networks, resulting in a simulated agent that achieved collision-free motions and human-like lane changes after a few learning rounds.
We propose an inverse reinforcement learning (IRL) approach using Deep Q-Networks to extract the rewards in problems with large state spaces. We evaluate the performance of this approach in a simulation-based autonomous driving scenario. Our results resemble the intuitive relation between the reward function and readings of distance sensors mounted at different poses on the car. We also show that, after a few learning rounds, our simulated agent generates collision-free motions and performs human-like lane change behaviour.