A Decision-Making Strategy for Vehicle Autonomous Braking in Emergency via Deep Reinforcement Learning

Autonomous braking through vehicle precise decision-making and control to reduce accidents is a key issue, especially in the early diffusion phase of autonomous vehicle development. This paper proposes a deep reinforcement learning (DRL)-based autonomous braking decision-making strategy in an emergency situation. Three key influencing factors, including efficiency, accuracy and passengers' comfort, are fully considered and satisfied by the proposed strategy. First, the vehicle lane-changing process and the braking process are analyzed in detail, which include the critical factors in the design of the autonomous braking strategy. Second, we propose a DRL process that determines the optimal strategy for autonomous braking. Particularly, a multi-objective reward function is designed, which can compromise the rewards achieved of different brake moments, the degree of the accident, and the comfort of the passenger. Third, a typical actor-critic (AC) algorithm named deep deterministic policy gradient (DDPG) is adopted for solving the autonomous braking problem, which can improve the efficiency of the optimal strategy and be stable in continuous control tasks. Once the strategy is well trained, the vehicle can automatically take optimal braking behavior in an emergency to improve driving safety. Extensive simulations validate the effectiveness and efficiency of our proposal in terms of learning effectiveness, decision-making accuracy and driving safety.

I. Introduction

Achieving driving safety is one of the top priorities that traffic participants and intelligent transportation systems (ITS) should pursue [1]. Statistics show that about 90% of traffic accidents are caused by driver errors [2]. On one hand, the driver’s distraction, misjudgment, and misoperation while driving will increase the risk of accidents. According to research results, in most serious rear-end collisions, the driver of the following vehicle (FV) usually takes incomplete braking or does not take effective braking [3]. On the other hand, sudden lane-changing or braking of the leading vehicle (LV) will cause the FV to have insufficient time to take appropriate measures, resulting in collision accidents. In addition to developing accurate risk assessments and timely warnings for collision avoidance systems, there is a need to develop connected and autonomous vehicles (CAVs) to achieve true autonomy.