基于模糊PID控制的车辆纵向优化CACC系统

发布时间：2018-10-29 19:25

【摘要】：随着智能交通系统ITS（Intelligent Transport System）框架的提出，汽车协同式自适应巡航控制系统CACC（CooperativeAdaptive Cruise Control）作为其重要的组成部分，在已经大范围普及的定速巡航控制CC（Cruise Control）系统和自适应巡航控制ACC（Adaptive Cruise Control）系统基础上，借助专用短程通信技术DSRC（Dedicated ShortRange Communications），实现了车与车之间的数据通信。通过与其他车辆进行实时的信息共享，获得本车和周围车辆的各项行驶数据，感知当前行驶状态，并针对不同的状态选择最适宜的驾驶行为来操纵汽车。相比于其他方法，CACC系统使车辆更加智能化，对车辆控制更加合理、准确，对道路突发状况响应迅速，从而能够大幅度提高交通系统的综合通行效率，并增强车上乘客的安全性和舒适性。本文设计了一种基于模糊PID控制的车辆纵向优化CACC控制系统。该系统能够实现在DSRC车车通信环境下，对于本车的纵向跟随控制。针对CACC系统中的两个主要功能——车车通信和本车控制，该系统结构可分为两层。第一层为感知层，主要负责接收通过DSRC端获得的周围车辆的行驶数据和本车雷达端获得的与前车距离数据。考虑到CACC系统中对数据的精度和实时性要求较高，而通常由车上传感器获得的数据精度偏低，并伴有一定延时，因此通过DSRC通信得到的其他车辆数据是无法直接使用的，所以本文创新地提出了一种加速度补偿平滑方法ACS（Acceleration CompensationSmoothing），通过设计一个2阶IIR滤波器和加速度补偿公式来对DSRC通信端接收到的数据进行优化，使其能够适应CACC系统要求，，并去掉DSRC通信中一些冗余数据，仅保留周围车辆运动学参数并与雷达端的距离数据整合，生成当前时刻状态模型。第二层为控制层，根据模糊PID控制理论，设计出一个针对车辆纵向行驶的CACC控制器，该控制器将感知层传来的状态模型作为输入，将两车速度参数作为主要计算量，其他参数作为辅助调节量，通过模糊控制动态调整PID调节器的各个参数，从而根据不同的行驶状态得到相应时刻优化的车辆的控制量（油门踏板或制动踏板的值），并输出到车辆动力学模型的执行机构对车辆进行控制，相比传统控制方法，有很强的适应性和稳定性。在Matlab/Simulink下建立相应的仿真程序，运用Matlab Filter工具箱和Fuzzy Logic工具箱等设计出符合CACC系统要求的通信模块、控制模块和车辆动力学模块，并借助Logitech G27方向盘搭建驾驶员模拟平台，进行仿真试验。试验结果表明，本文提出的感知层ACS方法能够提高DSRC通信接收端得到数据的精度和平滑性，该CACC模糊PID控制器可以在各种典型纵向工况中得到最优的数据输入，对车辆进行优化跟随控制，并能保证安全性和舒适性。
[Abstract]:With the development of intelligent transportation system (ITS (Intelligent Transport System) framework, CACC (CooperativeAdaptive Cruise Control) is an important part of automobile cooperative adaptive cruise control system. On the basis of CC (Cruise Control) system and adaptive cruise control ACC (Adaptive Cruise Control) system, which has been widely used, the data communication between vehicle and vehicle is realized by means of special short range communication technology (DSRC (Dedicated ShortRange Communications),). Through real-time information sharing with other vehicles, we can obtain the driving data of the vehicle and its surrounding vehicles, perceive the current driving state, and select the most suitable driving behavior to operate the vehicle according to the different states. Compared with other methods, CACC system can make vehicles more intelligent, more reasonable and accurate to control vehicles, and respond quickly to road emergencies, which can greatly improve the comprehensive traffic efficiency of traffic system. And enhance the safety and comfort of passengers on board. A vehicle longitudinal optimization CACC control system based on fuzzy PID control is designed in this paper. The system can realize the longitudinal following control of the vehicle under the DSRC vehicle communication environment. The system structure can be divided into two layers according to the two main functions of CACC system, that is, vehicle communication and local vehicle control. The first layer is the perceptual layer, which is mainly responsible for receiving the driving data of the surrounding vehicles obtained through the DSRC terminal and the distance data from the radar end of the vehicle to the front car. Considering the high requirement of precision and real time of data in CACC system, but the data precision obtained by vehicle sensor is on the low side and accompanied by certain delay, so other vehicle data obtained by DSRC communication can not be used directly. So this paper proposes an acceleration compensation smoothing method, ACS (Acceleration CompensationSmoothing), which optimizes the data received by DSRC communication terminal by designing a second-order IIR filter and acceleration compensation formula, so that it can meet the requirements of CACC system. Some redundant data in DSRC communication are removed and only the kinematics parameters of the surrounding vehicle are retained and integrated with the range data of the radar terminal to generate the current state model. The second layer is the control layer. According to the fuzzy PID control theory, a CACC controller is designed for the vehicle running longitudinally. The controller takes the state model from the perceptual layer as the input and the two vehicle speed parameters as the main calculation amount. The other parameters are used as auxiliary regulation, and the parameters of the PID regulator are dynamically adjusted by fuzzy control, so as to obtain the optimal control quantity of the vehicle (the value of the throttle pedal or brake pedal) according to the different driving conditions. The actuator output to the vehicle dynamics model controls the vehicle, which has strong adaptability and stability compared with the traditional control method. The corresponding simulation program is established under Matlab/Simulink. The communication module, control module and vehicle dynamics module are designed by using Matlab Filter toolbox and Fuzzy Logic toolbox, and the driver simulation platform is built with the help of Logitech G27 steering wheel. The simulation test was carried out. The experimental results show that the proposed perceptron layer ACS method can improve the accuracy and smoothness of the data obtained from the DSRC communication receiver, and the CACC fuzzy PID controller can obtain the optimal data input in various typical longitudinal conditions. The vehicle is optimized to follow control, and the safety and comfort can be guaranteed.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U463.6;U495;TP273.4

【参考文献】