Intelligent monitoring and residual analysis of tunnel point cloud data based on free-form approximation

Abstract

With the rapid development of urban rail transit and the aging of transportation infrastructure, the demand for shield tunnel disaster detection is about to break out. Most of the traditional monitoring methods require considerable manpower and time costs, which cannot satisfy the increasing requirements of tunnel operation and maintenance. Free-form model construction and deviation mechanism analysis are investigated in this paper to monitor the geometric deformation information of shield tunnels quickly and accurately. A method for identifying geometric features of tunnel sections based on the free-form B-spline approximation is proposed. The innovation of this paper lies in the intelligent recognition of common interference targets by the residual classification method. Furthermore, various Root Mean Squared Error (RMSE) distributions are investigated, which successfully realizes and verifies the clustering analysis of certain point cloud features.