A Deep Reinforcement Learning Approach to Eco-driving of Autonomous Vehicles Crossing a Signalized Intersection
by Joshua Ogbebor 1 
, Xiangyu Meng 1,* 
, Xihai Zhang 2
1 Division of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, 70803, United States of America
2 College of Electronic and Information, Northeast Agricultural University, Harbin, 150000, China
* Author to whom correspondence should be addressed.
Journal of Engineering Research and Sciences, Volume 1, Issue 5, Page # 25-33, 2022; DOI: 10.55708/js0105003
Keywords: reinforcement learning, eco-driving, connected vehicles, autonomous vehicles
Received: 26 February 2022, Revised: 10 April 2022, Accepted: 18 April 2022, Published Online: 12 May 2022
APA Style
Ogbebor, J., Meng, X., & Zhang, X. (2022). A Deep Reinforcement Learning Approach to Eco-driving of Autonomous Vehicles Crossing a Signalized Intersection. Journal of Engineering Research and Sciences, 1(5), 25–33. https://doi.org/10.55708/js0105003
Chicago/Turabian Style
Ogbebor, Joshua, Xiangyu Meng, and Xihai Zhang. “A Deep Reinforcement Learning Approach to Eco-driving of Autonomous Vehicles Crossing a Signalized Intersection.” Journal of Engineering Research and Sciences 1, no. 5 (May 2022): 25–33. https://doi.org/10.55708/js0105003.
IEEE Style
J. Ogbebor, X. Meng, and X. Zhang, “A Deep Reinforcement Learning Approach to Eco-driving of Autonomous Vehicles Crossing a Signalized Intersection,” Journal of Engineering Research and Sciences, vol. 1, no. 5, pp. 25–33, May 2022, doi: 10.55708/js0105003.
This paper outlines a method for obtaining the optimal control policy for an autonomous vehicle appro med that traffic signal phase and timing information can be made available to the autonomous vehicle as the vehicle approaches the traffic signal. Constraints on the vehicle’s speed and acceleration are considered and a microscopic fuel consumption model is considered. The objective is to minimize a weighted sum of the travel time and the fuel consumption. The problem is solved using the Deep Deterministic Policy Gradient algorithm under the reinforcement learning framework. First, the vehicle model, system constraints, and fuel consumption model are translated to the reinforcement learning framework, and the reward function is designed to guide the agent away from the system constraints and towards the optimum as defined by the objective function. The agent is then trained for different relative weights on the travel time and the fuel consumption, ults are presented. Several considerations for deploying such reinforcement learning-based ents are also discussed.
