Research And Implementation Of Forward Active Collision Avoidance System For Intelligent Vehicle Based On LiDAR

With the development of automotive industry and embedded technology,smart car has gradually become an important development direction of future automobiles.Active collision avoidance system,as an important part of Advanced Driver Assistant System(ADAS)and self-driving car,can greatly improve the driving safety.In this thesis,The active collision avoidance system was taken as the research object,the hardware and software architecture of the system was constructed,the obstacle detection algorithm based on LiDAR was designed,the active anti-collision strategy and open-loop control strategy was developed,and real vehicle and simulation experiment was carried out to verify the feasibility of the active collision avoidance system.The main contents are as follows:The hardware and software architecture of active collision avoidance system was constructed.In terms of hardware,the sensing system based on LiDAR was designed,The industrial computer was used as the controller of system,and the CAN communication circle was used as the hardware device to realize the communication between the industrial computer and the vehicle.In terms of software,a loosely-coupled software framework was built based on the basic principle of Robot Operating System(ROS).The obstacle detection algorithm based on LiDAR was designed.The algorithm completed the transformation of coordinate system,achieved the extraction of region of interest and the removal of abnormal points by height filter,body point filter and voxel filter.Based on Ray Ground Filter,An optimization algorithm was proposed to segment the ground points,and point clouds was clustered by optimizing Euclidean clustering algorithm.Finally,the centroid,outline and distance information of the obstacle were obtained,and were sent to the vehicle safety decision module by using topic communication mechanism of ROS.The active anti-collision strategy and the open-loop control strategy were developed.Based on the basic principles of braking process analysis and the characteristics of active collision avoidance system,the safe distance model with parameters such as relative speed was established.A relative speed acquisition method based on time stamp mechanism of ROS system was proposed.Finally,according to the open-loop control algorithm,the collision avoidance process was controlled by real-time acquisition of model parameters.In order to verify the feasibility of this scheme,an intelligent vehicle experimental platform was set up for experimental verification.In these experiment,the obstacle detection effect test and obstacle detection distance accuracy test were carried out respectively,and active collision avoidance system function was verified by Car Sim simulation and real vehicle experiment.The experimental results show that the forward active collision avoidance system based on LiDAR is feasible.