汽车工程 ›› 2023, Vol. 45 ›› Issue (5): 719-734.doi: 10.19562/j.chinasae.qcgc.2023.05.002

所属专题: 智能网联汽车技术专题-控制2023年

• • 上一篇    下一篇

融合预瞄特性的智能电动汽车稳定性模型预测控制研究

贺伊琳1,马建1(),杨舒凯2,郑威1,薛启帆1   

  1. 1.长安大学汽车学院,西安 710064
    2.西安科技大学电气与控制工程学院,西安 710064
  • 收稿日期:2022-11-07 修回日期:2022-12-05 出版日期:2023-05-25 发布日期:2023-05-26
  • 通讯作者: 马建 E-mail:majian@chd.edu.cn
  • 基金资助:
    国家自然科学基金(52172362);国家重点研发计划项目(2021YFB2501201)

Research on Stability Model Predictive Control of Intelligent Electric Vehicle with Preview Characteristics

Yilin He1,Jian Ma1(),Shukai Yang2,Wei Zheng1,Qifan Xue1   

  1. 1.School of Automobile,Chang’an University,Xi’an 710064
    2.School of Electrical and Control Engineering,Xi’an University of Science and Technology,Xi’an 710064
  • Received:2022-11-07 Revised:2022-12-05 Online:2023-05-25 Published:2023-05-26
  • Contact: Jian Ma E-mail:majian@chd.edu.cn

摘要:

提出了一种融合预瞄特性的智能电动汽车稳定性前馈+反馈控制方法。建立车辆预瞄模型,通过汽车在环境感知时的前视行为,引入道路曲率作为车辆动力学特性的影响因素。基于在前视信息指导下的车辆位姿变化,根据道路附着能力和车速指数模型描述期望纵向车速,建立轮胎侧偏刚度补偿的稳定性前馈控制方法。同时,采用模型预测控制(MPC)设计稳定性反馈控制律,根据车辆的预瞄信息自适应调整预测模型参数和预测时间,消除前馈控制误差及路面扰动等不确定性因素带来的影响。研究结果表明,本文提出的控制策略下汽车质心侧偏角、横摆角速度和侧向加速度小,且跟踪精度更高。仿真试验中,相比于无控制、MPC反馈控制与前馈+定参数MPC反馈控制,本文提出的控制策略在双移线工况1下质心侧偏角峰值分别减小了41.3%、28.9%和10.0%,横摆角速度峰值分别减小了18.0%、6.7%和2.0%,双移线工况2下质心侧偏角峰值分别减小了27.2%、8.7%和8.0%,横摆角速度峰值分别减小了16.9%、12.9%和8.6%;相比于MPC反馈控制与前馈+定参数MPC反馈控制,在蛇行工况1下,质心侧偏角峰值分别减小了49.8%与34.8%,横摆角速度峰值分别减小了21.8%与12.7%;在蛇行工况2下,质心侧偏角峰值分别减小了36.6%和18.6%,横摆角速度峰值分别减小了17.7%和12.4%。

关键词: 智能电动汽车, 稳定性控制, 预瞄特性, 前馈+反馈控制, 模型预测控制

Abstract:

A feedforward and feedback control method with preview characteristics for intelligent electric vehicle stability is proposed. The vehicle preview model is established, and the road curvature is introduced in as a factor influencing vehicle dynamic characteristics according to vehicle preview perception of the environment. Based on the change of vehicle posture guided by the preview information, the desired longitudinal vehicle speed is described according to the road friction condition and the vehicle speed index model and the stability feedforward control method for tire lateral stiffness compensation is established. At the same time, the model prediction control (MPC) is used to design the stability feedback control law, with the preview model and preview time adjusted according to the road information adaptively to eliminate the influence of uncertain factors such as feedforward control error and road disturbance. The research results suggest that the proposed control method has lower vehicle centroid sideslip angle, yaw rate and lateral acceleration, and higher tracking accuracy. In simulation test, compared with no control method, MPC feedback control, and feedforward + MPC feedback control with fixed parameter, the control strategy proposed in this paper reduces the peak mass center sideslip angle by 41.3%, 28.9% and 10.0% respectively under dual shift conditions, and the peak yaw rate by 18.0%, 6.7%, and 2.0% respectively. Under the other dual shift conditions, the peak values of the center of mass sideslip angle decreases by 27.2%, 8.7% and 8.0% respectively, and the peak value of the yaw rate decreases by 16.9%, 12.9% and 8.6% respectively. Compared with MPC feedback control, feedforward + MPC feedback control with fixed parameter, the maximum vehicle sideslip angle decreases by 49.8% and 34.8% respectively, and the maximum yaw rate decreases by 21.8% and 12.7% respectively under the S-shaped condition. Under the other S-shaped condition, the maximum vehicle sideslip angle decreases by 36.6% and 18.6%, and the maximum yaw rate decreases by 17.7% and 12.4% respectively.

Key words: intelligent electric vehicle, stability control, preview characteristics, feedforward + feedback control, model predictive control