| The autonomous vehicle is one kind of mobile robots, whose key technologies are related to frontier research fields of environmental perception, pattern recognition, navigation and positioning, intelligent decision-making, control engineering, computer technology, and many other disciplines. It is not only the best experimental platform to exam the cognitive theory and the critical technology, but also, at the same time, its unit technology can help humans drive cars automatically to improve the driving safety and efficiency.In the technology of autonomous vehicle, the high speed and high precision self-adaptative control of trajectory tracking in unknown traffic conditions is always the key difficulty. To deal with the problems of autonomous vehicle’s motion control in urban environment and to improve the known traditional method, this paper presents some new solutions dealing with the specific characteristics of vehicle control model and complex diversity of city environment, to enhance the adaptability of the algorithms. Our research includes the following several aspects, specifically:1) We first briefly introduce the concept of vehicle control theory and the state-of-the-art international and domestic achievements. This includes the understanding of vehicle’s kinetic characters with its modeling and simulation, the understanding of the environmental influence on the vehicle’s and the understanding of the driving process upon considering the way that human drives.2) This paper also summarizes the classification of urban streets and their respective characteristic, in a way that the tasks that the autonomous vehicle faced in various conditions clear. With the so-called "Intelligent Pioneer" as a basis platform of the autonomous vehicle, we introduce the platform of sensing systems, decision-making systems, control systems, and implementing agencies. We propose a control system’s research strategy and design idea that combine the key problems of autonomous vehicle’s motion control in urban environment.3) We analyze the vehicle longitudinal dynamics and study the method of vehicle longitudinal control. The power transmission characteristics and brake mechanical properties of "Intelligent Pioneer" is analyzed and summarized, according to the behavior of driver’s speed control, combining the expert control method. We develop the expert PID control algorithm, that establishes expert rules according to the driving experience, solves the problem of control accuracy when the transmission system is highly nonlinear and the longitudinal interference is too complex, so that the robustness of speed control system can be improved and the goal of speed control of autonomous vehicle’s driving in urban road can be established.4) We study the method of lateral control for vehicle operation dynamics. The model of vehicle operation dynamics is established for the research of vehicle operation dynamics and tire lateral mechanics. Upon the formally established theory and traditional PID control, this paper presents a new autonomous vehicle’s lateral control algorithm based on a composite of the CMAC and PID control. This enables the system to compensate automatically when the model and input signal occur changes unpredictability, such that the autonomous vehicle can drive steadily and accurately in any kind of urban environments.5) We study the kinematic characteristics when a car travels at a high speed. We consider the restricted relationship between vertical control and lateral control federatively and establish the constraint between the highest speed and biggest steering angle at the request of stability and comfort, to ensure the safety of the vehicles.At last, this paper takes "Intelligent Pioneer" as the experimental platform to check the feasibility and to examine the research above in the real urban environment. We also summarize the advantages and shortages of control system’s design procedures. Future research directions are also proposed. |
PDF Full Text Request