Turning Radius Prediction Method for Tracked Vehicles Based on PSO-BP Algorithm

Abstract

Crawler vehicles always slipped during the steering process. To address this problem, this paper uses particle swarm algorithm (PSO) to optimize the initial weights and thresholds of the BP neural network and establishes a turning radius prediction model based on the PSO-BP neural network. The model takes the turning angle as the input and the turning radius as the output. Kalman filter is used for data processing to eliminate random errors during the test process. The law between the physical parameters and algorithm parameters in the model is discussed by changing the range of turning angle and the number of hidden layers and initialization populations, and the reliability of the model is verified by a real vehicle test. The results show that it is feasible to predict the turning radius in the presence of slip by using the PSO- BP neural network algorithm, and the accuracy of the prediction model can reach 99% after Kalman filtering. The prediction model of the turning radius proposed in this paper provides a certain reference for the prediction of the turning radius of tracked vehicles under actual conditions.