Deep Reinforcement Learning for Intelligent Transportation Systems: A\n Survey

Abstract

Latest technological improvements increased the quality of transportation.\nNew data-driven approaches bring out a new research direction for all\ncontrol-based systems, e.g., in transportation, robotics, IoT and power\nsystems. Combining data-driven applications with transportation systems plays a\nkey role in recent transportation applications. In this paper, the latest deep\nreinforcement learning (RL) based traffic control applications are surveyed.\nSpecifically, traffic signal control (TSC) applications based on (deep) RL,\nwhich have been studied extensively in the literature, are discussed in detail.\nDifferent problem formulations, RL parameters, and simulation environments for\nTSC are discussed comprehensively. In the literature, there are also several\nautonomous driving applications studied with deep RL models. Our survey\nextensively summarizes existing works in this field by categorizing them with\nrespect to application types, control models and studied algorithms. In the\nend, we discuss the challenges and open questions regarding deep RL-based\ntransportation applications.\n