Intelligent Vehicle Path Tracking Under The Known And Unknown Lateral Velocity Circumstances

With rapid increase of vehicle penetration, various traffic problems occur frequently, which results in lots of trouble to people’s work and life and a large number of casualties and economic losses. Due to occurrence of the frequent traffic problems, many researchers begin to pay close attention to intelligent vehicle control. Intelligent vehicle is a kind of typical syntheses of advanced technologies, which uses synthetically various technologies including computer, robotics, sensors, controllers, and information communication technologies, and has a broad prospect in many engineering fields. Path tracking control is one of the key technologies in the field of intelligent vehicle control. It is the key to ensure vehicles run along the desired path accurately and safely. However, the fundamental problem of vehicle’s precise tracking path is lateral control, which requires a higher performance of the controller that makes an intelligent vehicle has a better performance when deals with some emergent events. The vehicle lateral velocity is the key factor to determine whether vehicles travel in the desired route precisely in the driving process. When the lateral displacement appears, a controller is required to adjust to make vehicles come back to safe state in order to realize driving safely. At present, the control technology of lateral velocity is comparatively mature, but the most research works are based on the condition of known lateral velocity. In practice, the lateral velocity can be measured by a sensor, which is with high cost. Thus, it’s expensive to use this kind of sensors in the commercial vehicle field. Under the vehicle lateral velocity is unknown, we design an extended state observer, which can estimate the vehicle lateral velocity, and the controller with the extended state observer can achieve a path tracking control task. In this way, the cost of the control system can be reduced while the vehicle safety is ensured. Those motivate our study in this thesis.The main works in this thesis are as follows:Chapter 1 mainly introduces the research background and significance of intelligent vehicles, and the development status of the vehicle path tracking control in recent years is recalled.Chapter 2 mainly studies the path tracking problem of vehicles when the lateral velocity is known. For the case that the lateral velocity of intelligent vehicle is known, a polyhedron linear vehicle system model with time-varying uncertainty cornering stiffness is established, and a corresponding controller is put forward to realize vehicle path tracking problem. Sufficient condition for the existence of controller gain is derived through Lyapunov method and linear matrix inequalities (LMI). Finally, the validity of the proposed method is demonstrated through the simulation experiments.Chapter 3 mainly investigates the path tracking problem of vehicles when the lateral velocity is unknown. For the case that the lateral velocity of intelligent vehicle is unknown, a vehicle model with unknown disturbance is established, and an extended state observer is constructed to estimate the unknown velocity and disturbance. Base on the extended state observer, active disturbance rejection controller is designed to realize vehicle path tracking control. The effectiveness of extended state observer and the anti-disturbance capability of active disturbance rejection controller is demonstrated by simulation experiment.Chapter 4 the works of this thesis are summarized, and the future work of the proposed control methods are prospected.