Research On Road Information Extraction Method Based On Laser Radar

The ride comfort of the vehicle is closely related to the performance of the vehicle suspension.At present,the vehicle suspension can only passively buffer the force transmitted by the road surface to the body to keep the vehicle stable.If the terrain in front of the vehicle can be measured online and at high speed by optical sensor,the vehicle suspension can adjust the stretching of each suspension according to the terrain information in front of the vehicle to adjust and control the vehicle position and attitude,so that the vehicle can maintain good maneuverability and comfort at high speed.Based on the national key research and development plan "Research on special chassis and suspension key technology of high-powered emergency rescue vehicles(including fire vehicles)",this paper studies how to conduct online high-speed measurement of terrain in front of vehicles through laser radar and other sensors,mainly focusing on two aspects of ground segmentation and obstacle detection.First of all,the principle and application of vehicle borne laser scanning technology are introduced briefly,and a vehicle front terrain scanning experiment platform is designed and built,which contains laser radar,inertial measurement unit,global positioning system and other sensors for vehicle front terrain scanning experiment and experimental data acquisition.Then,aiming at the problem that the current ground segmentation algorithm does not consider the small angle tilt of the ground point cloud in the radar coordinate system,a ground segmentation algorithm based on the point cloud characteristic line and plane calibration is studied.Firstly,the point cloud is compressed,and then the reference of ray ground is used Filter algorithm,analyze the relationship between the height difference between adjacent points and the height threshold,perform rough segmentation of ground and non ground points,and finally finish the fine segmentation of ground non ground through plane calibration.The experimental results show that the algorithm has a great advantage in reducing the ground level error.Finally,aiming at the problem that the current obstacle detection algorithm can notdistinguish the positive and negative obstacles and the influence of non road elements,this paper uses the clustering idea of Euclidean algorithm for reference,and studies an obstacle detection algorithm based on improved Euclidean clustering on the basis of the point cloud ground segmentation algorithm studied in this paper,which effectively solves the influence of non road elements on obstacle detection,The positive and negative obstacles are extracted and the information is identified.At the same time,the effectiveness of the algorithm is verified by experiments.