Deep Reinforcement Learning for Intelligent Transportation Systems: A Survey

Ammar Haydari,Yasin Yilmaz

2020-05-03

Abstract:Latest technological improvements increased the quality of transportation. New data-driven approaches bring out a new research direction for all control-based systems, e.g., in transportation, robotics, IoT and power systems. Combining data-driven applications with transportation systems plays a key role in recent transportation applications. In this paper, the latest deep reinforcement learning (RL) based traffic control applications are surveyed. Specifically, traffic signal control (TSC) applications based on (deep) RL, which have been studied extensively in the literature, are discussed in detail. Different problem formulations, RL parameters, and simulation environments for TSC are discussed comprehensively. In the literature, there are also several autonomous driving applications studied with deep RL models. Our survey extensively summarizes existing works in this field by categorizing them with respect to application types, control models and studied algorithms. In the end, we discuss the challenges and open questions regarding deep RL-based transportation applications.

Machine Learning,Multiagent Systems,Signal Processing,Systems and Control

What problem does this paper attempt to address?

The paper attempts to address the application of Deep Reinforcement Learning (DRL) in Intelligent Transportation Systems (ITS) to improve the efficiency of traffic control and management. Specifically, the paper focuses on the following aspects: 1. **Traffic Signal Control (TSC)**: How to use deep reinforcement learning to optimize traffic signal control strategies to reduce traffic congestion, improve road utilization, and reduce emissions. 2. **Autonomous Driving**: Exploring how deep reinforcement learning can be used to achieve autonomous driving of vehicles, including key issues such as path planning, obstacle avoidance, and decision-making. 3. **Multi-Agent Systems**: In traffic systems, how multiple agents (such as multiple traffic signals or multiple autonomous vehicles) can work together to achieve globally optimal traffic management. 4. **Challenges and Future Research Directions**: Discussing the main challenges currently faced by the application of deep reinforcement learning in traffic systems, such as the stability and scalability of algorithms, and future research directions. Overall, the paper aims to summarize the current state of the application of deep reinforcement learning in intelligent transportation systems through a review of existing literature, identify existing problems, and provide directions for future research.

Deep Reinforcement Learning for Intelligent Transportation Systems: A Survey

Deep Reinforcement Learning for Intelligent Transportation Systems: A Survey

Uniformity of Markov Elements in Deep Reinforcement Learning for Traffic Signal Control

Reinforcement learning: Introduction to theory and potential for transport applications

A survey on deep reinforcement learning approaches for traffic signal control

Recent Advances in Reinforcement Learning for Traffic Signal Control

Leveraging reinforcement learning for dynamic traffic control: A survey and challenges for field implementation

Enhancing Transportation Systems Via Deep Learning: A Survey

Leveraging Deep Reinforcement Learning for Traffic Engineering: A Survey

Deep Reinforcement Learning and Transportation Research: A Comprehensive Review

Reinforcement Learning in Urban Network Traffic Signal Control: A Systematic Literature Review

Deep Reinforcement Learning for Autonomous Driving: A Survey

A bibliometric analysis and review on reinforcement learning for transportation applications

Deep reinforcement learning for traffic signal control with consistent state and reward design approach

A Survey on Deep Learning-Based Traffic Signal Control

A Survey on Reinforcement Learning Models and Algorithms for Traffic Signal Control

A Deep Reinforcement Learning Approach to Traffic Signal Control With Temporal Traffic Pattern Mining

Deep Reinforcement Learning for Adaptive Traffic Signal Control

Deep Reinforcement Learning for Traffic Light Control in Intelligent Transportation Systems

Towards Robust Deep Reinforcement Learning for Traffic Signal Control: Demand Surges, Incidents and Sensor Failures

Adaptive Traffic Signal Control With Deep Reinforcement Learning and High Dimensional Sensory Inputs: Case Study and Comprehensive Sensitivity Analyses