Smart Car To Decoupling Control Method Of Research

With the rapid development of scientific and technological progress and the world economy in the past two decades, due to traffic congestion and the frequent occurrence of the accidents, prompting us to study Intelligent Vehicle and Automated Highway System. To some extents, the techniques can improve vehicle safety and maneuverability.The related research with Intelligent Vehicle can be further divided into three directions:(1) lateral control;(2) longitudinal control;(3) the combined control. Lateral control usually refers to the vehicle steering control, directly affecting the performance of cornering operation and lane change operation, so it is necessary to design a good steering controller. On the other hand, the independent control of the vehicle lateral dynamics and yaw dynamics can improves vehicle cornering ability. This paper proposes a decoupling control scheme based on mixed sensitivity method and also takes into account both the coupling and uncertainty effect on the system.The main results in the dissertation are as follows:In chapter1, we mainly introduce the research background, research situation at home and abroad, and the existing problems in the field.In chapter2, vehicle is a multivariable and highly coupled control system, so it is impossible to use a precise mathematical model to characterize. This chapter shows six degrees of freedom model of vehicle with Newton’s laws. Although the accuracy of the model is relatively high, but taking into account the actual calculation of costs, difficult to realize, in the case of the five basic assumptions, we provide a single track vehicle model with two degrees of freedom.In chapter3, for the vehicle steering decoupling problem, we propose a H∞mixed sensitivity decoupling scheme based on model tracking. We skillfully combine model tracking and decoupling by selecting a diagonal reference model, to achieve the purpose of decoupling by reducing the tracking error. Then we transfer the designed control problem into a standard H∞problem with mixed sensitivity method. A robust controller is obtained by solving a set of linear matrix equalities.Finally, we do the experiments in the case of zero desired yaw rate and zero desired slip angle respectively, the results show that vehicle has better decoupling performance.In chapter4, taking into account the slip angle information used in the third chapter is very difficult to obtain. In this chapter we design a luenberger observer to estimate the value of the slip angle. We propose an observer-based controller with mixed sensitivity method. In the simulation, the slip angle value estimated by the observer is compared with the exact value of the slip angle provided by the Carsim software, the results demonstrate the effectiveness of the proposed method.In chapter5, the main results of the thesis are concluded, and some research issues in future are proposed.