Research On Pavement Flatness Evaluation Algorithm Based On Vehicle 3D Laser Point Cloud Data

Pavement flatness is one of the main technical indicators for evaluating the quality of road pavement,which is related to the safety,comfort and service life of the vehicle.Most of the traditional pavement flatness evaluation methods are analyzed by measuring the discrete elevation information on the road longitudinal section in the field.There are shortcomings such as large manpower and material consumption and low accuracy of analysis results.In this paper,the 3D laser point cloud data collected by the vehicle mobile measurement system has been fully studied and considered.Based on the research on the evaluation method of pavement flatness at home and abroad,this paper proposes an algorithm for extracting the required road profile elevation information from the laser point cloud data to evaluate the pavement flatness,and obtains the expected evaluation results.The main research contents include these:(1)The algorithm of pavement's laser point cloud filtering is studied.Before the pavement flatness evaluation,it is necessary to filter the laser point cloud of the pavement to improve the calculation efficiency and accuracy of the algorithm.Compare the filtering principle of several point cloud filtering algorithms,and finally adopt the morphological filtering method suitable for the data characteristics of this paper.Taking the road elevation at the track point of the synchronously recorded point cloud data acquisition as the reference,setting the height difference threshold to filter the noise points such as flying points,vehicle tire points and building points stored in the point cloud data to ensure the accuracy of the evaluation results.(2)The algorithm of interpolation point elevation of road profile is studied.The evaluation of pavement flatness is to analyze and evaluate the undulation of the longitudinal section of the road.The road longitudinal vector line is interpolated equidistantly by calling ArcGIS interface.Two kinds of interpolation point elevation algorithms are proposed based on the elevation information of the road point cloud,which are the interpolation point buffer analysis method and the laser point cloud rasterization method.By comparing and analyzing the retention results of the two algorithms' processing results on the detail of the road profile height difference,the interpolation point buffer analysis method is finally determined as the interpolation point elevation algorithm of this paper to generate the road profile line.(3)The algorithm of pavement flatness evaluation is studied.Referring to the traditional three-meter ruler method for pavement flatness evaluation method,the maximum depth value of each 10m section is extracted from the road longitudinal section line by the envelope method of setting the virtual ruler.Calculate standard deviation for all maximum depth values.And using the conversion relationship obtained by a large number of experiments conducted by the Highway Science Research Institute of the Ministry of Communications of China,the standard deviation is converted into the International Roughness Index.According to the International Roughness Index,the Riding Quality Index is calculated,and the accuracy of the evaluation results is verified by referring to the relevant specifications in China's Highway Technical Status Assessment Standard.(4)Verify algorithm reliability by setting up experiments.The envelope method used to obtain the maximum depth in this paper is borrowed from the three-meter ruler method.The operation principle of the two methods on the vertical section of the road is identical.It proves that this algorithm is theoretically feasible for flatness evaluation.Using the pavement flatness evaluation algorithm in this paper,the laser point cloud data of a highway and a secondary road in Shandong Province are processed.Compared with the results of the calculation of the pavement flatness calculation program published by the World Bank,it is concluded that the proposed algorithm has obtained the expected flatness evaluation results.