Pavement Information Detection Method Using Laser Radar On UAV

In the vehicle identification study,the use of fixed platform to install the sensor to collect data can not be applied to the emergency section of the situation.And the use of vehicles equipped with sensors will cause secondary damage to the road in rutting detection study.In this paper,we present a method of using multi-rotor unmanned aerial vehicle(UAV)to collect pavement information data by lidar.Then based on the lidar data,the vehicle identification algorithm and the rutting detection algorithm are studied.The main contents include the following aspects:(1)The current research of vehicle identification and rutting detection were summarized,the advantages and disadvantages of existing vehicle identification and rutting detection methods were analyzed,and the significance of using this method to study the vehicle identification algorithm and rutting algorithm was explained.(2)The design of the multi-rotor UAV equipped with the lidar acquisition system was proposed,which includes the hardware system scheme,the selection of the multi-rotor unmanned aerial vehicle and the laser radar.An experiment was designed to test the error of the lidar and describe the data characteristics of the lidar.Then,the preprocessing of lidar data was studied.(3)The vehicle identification algorithm based on the lidar data was studied.The vehicle identification algorithm consists of two parts.The first part is the vehicle type recognition algorithm.The cars with the vehicle height less than 3m were reclassified combining the existing vehicle classification and the characteristics of the lidar data.The algorithm was used to calculate the characteristic parameters of the vehicle.The other part is the calculation of the vehicle speed.The mathematical model of the vehicle length was established,and the calculation of the vehicle speed was calculated based on the vehicle length and the laser data.Finally,these two algorithms were verified by ground test.(4)Based on the rutting detection algorithm of LIDAR data,the existing rutting calculation index was analyzed,and the rutting depth is calculated by using virtual pull method.A vertical rutting calculation index is proposed to meet the characteristics of lidar data.Aiming at the characteristics of the lidar data of the Taibus block section of G207,the algorithm of rutting detection was studied.(5)Designed flight tests to use multi-rotor UAV equipped with lidar to collect pavement information data,including test methods,content,etc.Using the model identification and rutting detection algorithm studied in this paper,the lidar data collected by the flight tests were processed,and the results were analyzed.