Research On Campus Intelligent Bus Vehicle Obstacle Avoidance Control Strategy

Over the past three decades,with the rapid development of artificial intelligence,ambitious research has been carried out in the field of autonomous driving.Intelligent bus obstacle avoidance control is one of the important research directions of driverless technology,which provides a new way to solve the increasingly severe traffic problems,which makes the intelligent bus significantly improved in dealing with complex road traffic situations.The closer the study of smart buses is to the actual road conditions,the response of buses to dynamic obstacles and pedestrians,such as joining the fast traffic flow or avoiding other moving vehicles,should be considered.Therefore,this article carried on the thorough research to control the obstacle avoidance,the problem of obstacle avoidance control mainly includes the vehicle itself and the obstacle position detection and obstacle avoidance path generation,the choice of the optimal path and the accuracy of trajectory tracking,car body control decision of the car can be safely and successfully avoid obstacles,to avoid collision with the obstacles,has important significance to improve the intelligent bus security.Firstly,this paper expounds and analyzes the research status of intelligent vehicles at home and abroad,introduces in detail the global path planning and local path planning methods in path planning,and determines the path planning method in this paper.After introducing the key technologies related to the intelligent car,the kinematics and dynamics modeling of the intelligent car were studied,and the whole bus model composed of the vehicle kinematics model,the dynamics model and the magic formula tire model was established.Secondly,based on the analysis of the environment perception system of smart cars and buses,the GPS sensor is built,and the precise positioning of smart buses is realized by means of dead reckoning.Structures,millimeter wave radar sensor,the error because the sensor noise,so the Kalman Filtering model is set up,the original data filtering and estimation,in order to quickly and accurately obtain information on the intelligent bus around obstacles(relative distance,relative velocity and Angle),etc.,as an intelligent bus obstacle recognition,path planning and trajectory tracking,and other functions to provide accurate information.Based on rapid prototype development controller(d SPACE),an intelligent bus test platform was built,and real-time communication between the algorithm model and the intelligent bus platform was realized by CAN bus.Because the traditional path planning algorithm has some problems such as local optimality,large computational load of the model and even no solution of the model,a method of path and speed simultaneous planning based on the Lattice algorithm is proposed.According to the state of the vehicle,pick points in position space and time,and connect the starting state and the ending state by the fifth order polynomial,so as to obtain the planned horizontal and vertical trajectories.At the mining point,the feasible region is selected mainly according to the obstacles,and the constraint conditions of vehicle dynamic performance are considered to improve the efficiency of the algorithm and the solution space,so as to achieve better performance.The LQR algorithm is used to design the transverse controller and calculate the steering wheel Angle for body control.Finally,the obstacle avoidance control algorithm is tested on the intelligent bus platform.The security and real-time performance of the algorithm are analyzed.The results show that the obstacle avoidance trajectory designed by the proposed Lattice algorithm is smoother,and the LQR controller designed can accurately control the vehicle body,so that the intelligent bus can avoid the obstacles in front of it safely and stably.