Improved Image And Position Of Road Defect By 3D Ground Penetrating Radar

The urban roads are the lifeline of the city,but in recent years,the road collapsed accidents caused by underground concealed defects(cavity,non-compacting,debonding,etc.)posing a great threat to the traffic safety of citizens.Therefore,it is extremely urgent to apply appropriate geophysical methods for road detection and repair the defects detected in time to prevent collapse.Multi-channel 3D ground penetrating radar(GPR)is the most suitable non-destructive testing method for urban road detection currently.But the inaccurate positioning of the global positioning system(GPS),the channel imbalance,low recognition accuracy,etc.have affected the efficiency and accuracy of multi-channel 3D GPR in detecting urban roads.The main research contents are as follows:(1)To deal with the impacts of GPS errors on positioning accuracy of Multi-channel 3D GPR system,an analysis of the GPS errors' is carried out.After that,an azimuth correction algorithm is proposed and applied to continuous GPS positions in a field experiment in this paper.The results show that the GPS points around GPS errors have an azimuth abrupt change,and the azimuth correction algorithm is more effective than the Kalman filtering algorithm for the correction of GPS errors.(2)To compensate for the channel imbalance of the multi-channel GPR,this paper proposes an improved spectral whitening data processing method and applies it to an urban road underground cavity.The results show that this processing method can improve the recognition accuracy while detecting road defect.(3)To improve the accuracy of the defect detection in urban road,this paper carries out simulation experiments and field experiments.Based on the time-domain finite difference(TDFD)method,the numerical simulation models of typical urban road defects are established,and the 3D radar simulation results are obtained.The 3D radar image characteristics of the cavity,non-compacting,debonding,and common interference targets are summarized.The study of this paper can improve the accuracy of positioning and defect recognition of multi-channel 3D GPR in urban road detection,which helps efficient and highly accurate detection of urban road defects and prevent collapse from happening.