Highway Lane Change Decision-Making via Attention-Based Deep Reinforcement Learning
Jun‐Jie Wang, Qichao Zhang, Dongbin Zhao
Abstract
Deep reinforcement learning (DRL), combining the perception capability of deep learning (DL) and the decision-making capability of reinforcement learning (RL) [1], has been widely investigated for autonomous driving decision-making tasks. In this letter, we would like to discuss the impact of different types of state input on the performance of DRL-based lane change decision-making.
Topics & Concepts
Reinforcement learningReinforcementComputer scienceArtificial intelligencePsychologySocial psychologyAutonomous Vehicle Technology and SafetyTraffic control and managementAnomaly Detection Techniques and Applications