| Safety is always one of the most important topics in vehicle engineering. In recent years with the rapid development of transportation, traffic accidents of highway, especially the malignant traffic accident, such as the rear-end collision etc., are greatly increasing. Safety problem in transportation is increasingly outstanding. And Intelligent Vehicle (IV) emerges with the tide of times. IV is a complicated system that equipped with advanced technology such as electronics, computer, machinery etc. It is on the top level in the research field of vehicle engineering. It is involved in the combination of multiple academic subjects and integration of multiple high technologies and represents the development direction of automobile technology. The safety guarantee technology is the essential and indispensable component of IV research. The western developed countries, such as the U.S., U.K., Japan, etc., have launched their research programs to develop the safety guarantee technology of IV and automobile assistant safety driving and have already achieved great progress. However, because of the limited conditions of economy and scientific study in our country, the study of intelligent vehicle falls greatly behind the developed countries. Intelligent vehicle has been studied extensively these days. Lane tracking and platooning algorithms have been presented and tested rather extensively, yet autonomous lane changing and autonomous overtaking have not been fully studied. The main reason is the complexity of the lane change maneuver, since it incorporates lateral and longitudinal control in the presence of obstacles. In addition, the whole procedure should be completed quickly, smoothly and safely. Considering the condition of economy, industry and scientific research in our country, the paper has done some active research in the field of safety guarantee technology based on the vehicle active safety. The Cybercar is taken as the platform for experimental research, which is an autonomous model vehicle developed by IV Research Group of Jilin University. 1. Lane identification is the precondition of automated lane changing and overtaking. It can be divided into two parts: the lane detection and the lane tracking. l The algorithm of lane detection includes: digital image filtering, edge enhancement, binary-code image and Hough transformation. Firstly, Median value filter is used to eliminate or reduce the noise caused by the outside environment. Secondly, the edge information of lane image is enhanced by Sobel operator. And then based on the Otsu method, an effective algorithm is proposed to binary-code the lane image in order to segment and identify lane edge easily. Finally, Hough transformation is used to extract the feature of lane and detect the lane edge. Thus lane edge is detected and Area Of Interest (AOI) is established. l Based on the position correlation in the video images, the Kalman filter is introduced to predict the position of AOI to reduce the searching area. This algorithm integrates the Kalman filter and precognition of lane position, and improves the rapidity and accuracy of lane detection. 2. The trajectory planning is the purpose of automated lane changing and overtaking. Reasonable trajectory is the foundation of lane changing and overtaking for vehicles. l Using the minimum safety spacing (MSS) as an index of the safety lane changing, the kinematics of the vehicle involved in lane changing scenes was analyzed. The computation of MSS is correlated with the current vehicle velocity, the time reaching critical collision point, the relative velocity and relative acceleration between both vehicles to seek the conditions of vehicle collision avoidance. Using relation of triangle function, overtaking model without collision is developed. l On condition that there is no leading vehicle, the feasibility of motion planning based on sine function trajectory, circular arcs trajectory, trapezoid acceleration trajectory and polynomial trajectory is discussed. The advantages and disadvantages of the trajectory planning algorithms above are compared. And polynomial trajectory is selected as the motion trajectory of lane changing. Based on the previous algorithm, vehicle geometry model is set up to do further research on lane changing and overtaking without collision. In the end simulation environment based on VRML/Simulink language is established to verify the possibility of the method. 3. By analyzing the static characteristics of steering system and drivingsystem of Cybercar, the invert M sequence is chosen as input signal to identify the systems. The selection principle of the highest work frequency of system, clock period, sequence period and signal amplification is decided. At last the mathematics model for steering system and driving system can be obtained by using least square method. 4. The trajectory tracking is the realizing method of lane changing and overtaking. l The essence of lane changing and overtaking is matter of trajectory tracking in automated control realm. In terms of constraints of 'pure roll and no glide', the non-holonomic kinematics model of IV is built up. The 3-by-3 transformation matrix transforms the global posture errors to the local posture errors in the Cartesian coordinate system. l Based on the nonlinear state feedback control and backstepping method, the nonlinear controller with the virtual feedback parameters is designed to track the lane changing and overtaking trajectory predefined. Furthermore, the Lyapunov function is applied to discuss the closed loop stability of system and .the value scope of controller is confirmed. Finally, the simulation software based on Matlab\Simulink is developed. The simulation results show that the controller is reliable. In a word, the above systematic and scientific research work on automated lane changing and overtaking provides a valuable theoretical and technological support for the research and development of IV and the safety assistant driving system. It can be also taken as the reference material for the other research field such as the mobile robots and AGV etc. On the other hand, it can provide theory and technical support for the research on vehicle safety assistant driving in our country, thus it has obvious social and economical benefit.
|
PDF Full Text Request