Minimizing Energy Loss Decisions for Green Driving Platoon

Abstract

This paper presents the application of reinforcement learning (RL) in the vehicle communication simulation framework (Veins). Reinforcement learning methods for energy saving and greening in the field of autonomous driving have rarely been studied. Under a CACC platoon of green environmental protection, we investigate the use of reinforcement learning algorithms to train the behavior of member vehicles in the event of a serious collision in the front vehicle, so that platoon members can minimize collision damage and energy consumption from behavior which is not in line with the green theme. In terms of energy consumption metrics, the gradient policy algorithm has good convergence in computing the energy consumption problem. It is a feasible training decision planning algorithm for solving the minimization of energy consumption caused by decision behavior in platoon avoidance behavior.