A Digital Simulation for Bridges Management by Integrating LiDAR Scanning and Recurrent Neural Network with Long Short-Term Memory Blocks (RNN-LSTM)

Abstract

This paper aims specifically at developing an efficient but low-cost methodology that can help detect early transportation infrastructure damages either by permanent or periodic monitoring. In this research, we used LiDAR scanning units (ground units fixed on holders and movable units fixed on UAV) integrated with a novel deep neural network (DNN) for disease monitoring of bridges. The monitoring model is based on a recurrent neural network with long short-term memory blocks (RNN-LSTM) since the LiDAR scanning datasets have a time-dependent and memory-dependent behavior. The results give high performance, in this way, the monitoring of a real lifeline can be analyzed by combining with the data from Li-DAR and DNN models.