Full Vehicle Dynamic Model for PID Speed Controller Based on Practical Swarm Optimization Method

Abstract

The cruise control system CCS is considered to be an important system, because it is a fundamental component in autonomous vehicles. The CCS has high significant because CCS adds more autonomy to the vehicle. The CCS has many advantages such as making the driving less stressful and safer to the driver. In addition, it is important to the environment because it minimizes the fuel consumption and engine emissions. A full longitudinal vehicle dynamic model has been proposed in this work for forward motion. The model consist from drivetrain subsystem, resistance forces, and tire model. The significant in using full vehicle model gave a reality to the control performance specially tire model and slip ratio, and explained the vehicle behavior with/without the speed controller under different velocities and different conditions. The PID controller is being used for speed control system. The PID controller is widely used as its parameters are simple to adjust. Thus, many control systems especially controllers with feedback use PID controllers. Despite of its simple adjustment, it can be considered a very efficient controller, besides the system output response with PID controller become stable and reliable due its ability of linearization. The controller parameters which they are proportional KP, integral KI, and derivative KD are being tuned with different speeds for two times. The first by manual tuning and the second with using practical swarm optimization PSO algorithm. The overall performance of the proposed model has been evaluated with the two tuning methods at each desired speed. The experimental results show that PID based PSO gave better performance. With no overshot, zero steady state error, earliest raise time that 20 seconds earlier. This controller performance is desirable since it is difficult to adjust all the controller parameters at the same time. The PID based PSO Reaching the desired speed in a competitive time, furthermore it Reduce the instabilities that can be detected when the optimization is not used. The full model is being implemented in MATLAB/Simulink software environment