Research On Measurement Technique Of Road Elevation Information Front Of Vehicle Based On 3D-Laser Radar

The use of high mobility emergency rescue vehicles is mountain rescue.It often works on bad road conditions such as off-road.The characteristics and forms of the ground have great influence on vehicle ride comfort.Suspension is an important structure of the chassis part of the vehicle,which plays the role of transferring the force between the tire and the body and alleviating the impact of the road.The preview control active suspension adjusts the suspension according to the preview road elevation,which is achieved by mounting a laser sensor ahead of the vehicle to perceive the undulating condition of the road in advance.In this paper,we propose an approach to use 3D-lidar sensor?IMU and GPS to get accurate preview three dimensional point cloud datas of the road ? attitudes information of the vehicle and position information in order to extract the elevation information of the front road in advance.The approach is proposed according to the project national major R&D special subproject,"studies the method of recording the terrain in front of the car and the high speed data processing technology".The main work of this paper is as follows:Firstly,we established the data acquisition system based on 3D laser-sensor?IMU and GPS.The installation position?combination mode and basic working principle of each sensors of the system are introduced.At the same time,we analyzed the data format of the Ibeo-Lux2010 four line lidar.And the joint calibration of the laser radar and the inertial measurement unit is carried out on this basis.Secondly,we completed the datas collection.In this paper,multi-thread technology of boost library is used to control laser radar? IMU and GPS to collect data according to their sampling frequency,so as to get data needed for subsequent steps.The coordinate system for each sensors in the system is established and the error term of the system is analyzed.Thirdly,the data fusion of the lidar and the attitude position measuring device is completed.Because the error of the inertial measurement element accumulates with time,we design a SINS/GPS pine combined indirect Kalman filter to improve the measurement accuracy of the attitude angles.Then the three-dimensional coordinate information of the pavement in the geodetic coordinate system is obtained by coordinate transformation in MATLAB.Finally,we interpolate and fitting of these preview elevation information of the road by interpolation.First,we filter the point cloud datas by voxel grid filtering algorithm in order to down sample the point clouds.Second,we filter out road elevation datas on the width of the left and right tires by pass-through filtering algorithm,which is premise for extraction of road elevation information.In the end,we extract the key points from road elevation datas after pass-through filtering algorithm and interpolate and fitting of these key points to get the preview elevation information of the front road.