Simulated Autonomous Driving on Realistic Road Networks using Deep Reinforcement Learning

TL;DR

This paper introduces DSA^2, a software platform for testing deep reinforcement learning algorithms in realistic driving environments, and demonstrates its application in a vehicle speed regulation task on straight roads.

Contribution

The paper presents DSA^2, a new simulation tool for validating DRL algorithms in realistic driving scenarios, addressing a gap in existing research environments.

Findings

DSA^2 effectively simulates realistic road networks.

DRL algorithms can regulate vehicle speed according to different speed limits.

The platform enables testing in more typical driving conditions.

Abstract

Using Deep Reinforcement Learning (DRL) can be a promising approach to handle various tasks in the field of (simulated) autonomous driving. However, recent publications mainly consider learning in unusual driving environments. This paper presents Driving School for Autonomous Agents (DSA^2), a software for validating DRL algorithms in more usual driving environments based on artificial and realistic road networks. We also present the results of applying DSA^2 for handling the task of driving on a straight road while regulating the velocity of one vehicle according to different speed limits.