基于线性自抗扰控制的自动驾驶车辆纵向加速度控制算法研究

陈国迎 ¹ ，赵选铭 ² ，文良浒 ³ ，郑修磊 ²

1.吉林大学，汽车仿真与控制国家重点实验室

2.吉林大学，汽车工程学院

3.中国人民解放军63612部队

【摘要】 为提高自动驾驶车辆纵向加速度控制算法的鲁棒性，基于电动汽车纵向加速度系统仿真模型，提出了具有双观测器结构的线性自抗扰加速度跟踪控制算法，并通过带有遗忘因子的递归最小二乘法对阻力模型进行在线辨识，实现了驱动电机与制动系统之间的协调切换。通过搭建硬件在环 （ HIL ） 仿真平台对设计的控制算法进行了鲁棒性和实时性验证，结果表明，所提出的控制算法能够实现对加速度信号快速、稳定跟踪，能够在车辆纵向动力学模型未准确建模状态下对纵向加速度进行有效控制。

【关键词】 自动驾驶车辆，纵向加速度，线性自抗扰控制，递归最小二乘法，卡尔曼滤波

Research on Longitudinal Acceleration Control Algorithm of Autonomous Vehicle Based on Linear Active Disturbance Rejection Control

Chen Guoying ¹ ，Zhao Xuanming ² ，Wen Lianghu ³ ，Zheng Xiulei ²

1. State Key Laboratory of Automotive Simulation and Control ， Jilin University

2. College of Automotive Engineering ， Jilin University

3. Unit 63612 ， PLA

Abstract: In order to improve the robustness of the longitudinal acceleration control algorithm for autonomous vehicles，this paper proposes a linear self-rejecting acceleration tracking control algorithm with a double observer structure based on the affine model of the longitudinal acceleration system of electric vehicles，and the coordinated switching between the drive motor and the braking system is realized by online identification of the resistance model with a recursive least squares method with forgetting factor.The robustness and real-time performance of the designed control algorithm are verified by building a hardware-in-the-loop（HIL）simulation platform.The results show that the proposed control algorithm can achieve fast and stable tracking of the acceleration signal and can effectively control the longitudinal acceleration when the vehicle longitudinal dynamics model is not accurately modeled.

Key words: autonomous vehicle，longitudinal acceleration，linear active disturbance rejection control，recursive least square method，kalman filter tJTr+JP0xRMcoJs/k0E6tkzcbcn0tIqSsIAmdUUdbYCtfJZMcr7VSiT2tvxoUsGG