Optimization For Vehicles Longitudinal CACC System Based On Fuzzy PID Control

With the introduction of ITS (Intelligent Transport System) framework, the CooperativeAdaptive Cruise Control (CACC for short) system which is one of the most important partsof ITS can achieve the data communication between vehicles using Dedicated Short RangeCommunications (DSRC for short) technology, as the foundation of widely used CruiseControl and Adaptive Cruise Control. With the help of real time information sharing betweenother vehicles, CACC system obtains kinds of traveling data and gains the current drivingstate to choose an appropriate behavior to control the vehicle. Compared to other methods,CACC system makes vehicle more intelligent, vehicle control more reasonable and accurate.As a result, the efficiency of the transport system has been increased, and the safety andcomfort of passengers has been enhanced at the same time.This paper presents a framework for optimizing the vehicle longitudinal CACC systembased on fuzzy PID control. This framework enables the longitudinal following control of thevehicle with the support of DSRC vehicle-to-vehicle communication. Based on the two mainfunctions of CACC system, V2V communication and control of the vehicle, the structure ofthe framework is divided into two layers. The first layer is sensing layer, which is responsiblefor receiving driving data from DSRC and distance data from radar. Considering the fact thatthe CACC system requires highly accurate and real-timing data and that the vehicle sensorreceives poorly accurate and delayed data, the data received by DSRC is not readily availablefor use. Therefore, this paper presents an Acceleration Compensation Smoothing (ACS forshort) method, which optimizes data received by DSRC by implementing a2-order IIR andan acceleration compensation function. The method will qualify DSRC for the standardsimposed by the CACC system. It also will eliminates redundant data in DSRCcommunication, by retaining vehicle’s dynamic data and integrating the kinematics data withdata received by radar, thus developing a model for the current state. The second layer is control layer, which produces a longitudinal CACC controller based on fuzzy PID controltheory. The controller uses state model form sensing layer as input and the speed data ofvehicles as main calculating factor with other parameters as assistant data, and dynamicallyadjusts various PID controller parameters through fuzzy control, in order to produce theoutput of executive body to control the vehicle based on different state to gain the controlvolume (the quantity of accelerator or brake). The method produces more stable andadaptable results than traditional method.By developing a simulation under Matlab/Simulink and taking advantage of MatlabFilter toolbox and Fuzzy Logic toolbox, we are able to design the communication model、control model and vehicle dynamic model in accordance with the CACC system requirementsand carry out simulation with a help of Logitech G27. The simulation proves that the ACSmethod in sensing layer presented in this paper is, in fact, improving the accuracy andsmoothing of the data received by DSRC receiver. The experiment further shows that theCACC fuzzy PID controller produces optimal input data in various typical longitudinalconditions, and enables the following control while ensuring safety and comfort.