Rail track curve recognition model based on PSO-LSTM algorithm

In this study, a PSO-LSTM model is proposed[1]to increase the accuracy of rail curve detection. In order to alter the model's parameters, Particle Swarm Optimization (PSO) techniques are used to enhance the hidden layer and learning rate of the long-and-short-term memory neural network (LSTM). Afterward, apply the PSO-LSTM model to determine the rail's curvature. To make up the identifying process: Data processing, PSO algorithm optimization, and PSO-LSTM model prediction are the three steps. The PSO-LSTM[2]model is validated using data from train tracks, and it is contrasted with the Back Propagation(Bp)[3]and conventional LSTM[4]models. The outcomes demonstrate the superior accuracy and stability of the PSO-LSTM model.