An MPC-based asynchronous obstacle avoidance strategy for intelligent and connected vehicle platoon

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), Nov.2023, Vol.46 No.11: 1454-1459

DOI: 10.3969/j.issn.1003-5060.2023.11.003

Authors

PENG Liming $ ^{1} $, SUN Jun $ ^{1} $, WEI Zichun $ ^{2} $, BAI Xianxu $ ^{1} $

(1. School of Automobile and Traffic Engineering, Hefei University of Technology, Hefei 230009, China; 2. Viterbi School of Engineering, University of Southern California, Los Angeles 90089, USA)

Abstract and Keywords

Abstract: The effective motion planning for obstacle avoidance is the precondition to guarantee the driving safety of the intelligent and connected vehicle platoon. In this paper, an asynchronous obstacle avoidance strategy based on model predictive control (MPC) is proposed for intelligent and connected vehicle platoon to address the lateral control problem while the platoon is avoiding the obstacles. Firstly, the artificial potential field (APF) theory is used to establish the potential field of the surrounding traffic participants and roads. The desired lane and velocity information is obtained by the global planning module and the perception module. Secondly, comprehensively considering the APF, desired lane and velocity, and kinetic constraints, the MPC is used to accomplish the multi-objective control to avoid the obstacles. Besides, the asynchronous avoidance strategy is used for the platoon to decrease the driving risk and the lane space requirements. Finally, several obstacle avoidance scenarios are constructed based on the co-simulation of MATLAB/Simulink and TruckSim, including the single obstacle avoidance and multi-obstacle avoidance scenarios. The performances of the synchronous avoidance strategy and the asynchronous avoidance strategy are compared, and the result indicates that the MPC-based asynchronous avoidance strategy for intelligent and connected vehicle platoon can effectively improve the safety of the platoon system. Therefore, the feasibility and superiority of the proposed strategy is verified.

Keywords: intelligent and connected vehicle platoon; model predictive control(MPC); obstacle avoidance

作者信息

彭利明 $ ^{1} $，孙骏 $ ^{1} $，魏子淳 $ ^{2} $，白先旭 $ ^{1} $

(1. 合肥工业大学汽车与交通工程学院，安徽合肥 230009；2. 南加利福尼亚大学维特比工程学院，美国洛杉矶 90089)

摘要和关键词

摘要: 有效避障规划是保证智能网联车队行驶安全性的前提。文章提出一种基于模型预测控制(model predictive control, MPC)的智能网联车队异步避障策略,以解决智能网联车队规避障碍物时的横向控制问题。首先运用人工势场理论建立周围交通参与物和道路的势场,并由全局规划模块及感知模块获得目标车道和车速信息;其次基于模型预测控制器综合考虑人工势场、目标车道与车速以及车辆动力学限制进行多目标控制,以实现车队的避障规划,智能网联车队采用异步避障策略降低行车风险以及道路的占用面积;最后基于MATLAB/Simulink与TruckSim的联合仿真平台建立多个避障场景,包括单障碍物避障和多障碍物避障,对比车队的同步避障策略与异步避障策略的性能表现。结果表明,文章提出的基于MPC的智能网联车队异步避障策略能够有效提升车队行驶的安全性,证明该方法可行且优越。

关键词: 智能网联车队；模型预测控制（MPC）；避障；行驶安全；横向控制

Funding

安徽省新能源汽车创新工程资助项目（JZ2021AFKJ00002）

合肥工业大学学报(自然科学版)

Navigation