Estimating Vehicle-loss With Bp Neural Network And Genetic Algorithm

With the development of the automobile industry in China, localization R&D has been received more and more attention. Vehicle-loss is one of the most important input in the vehicle rig test. The requirements change frequently in the different R&D phase. The outdoor tests meet a great impact on the environment and weather factors, so that our human resource and time are not enough to finish the work. The aim of the issue is to bring out a method which is used to simulate the vehicle-loss model in different situation.The feasibility analysis of BP Neural Networks which is used to improve the vehicle-loss measurement is performed in this paper. Through the nntool of Matlab software to finish the networks design and combination of genetic algorithms which optimize the network weights and threshold, we can see that the network performance has been improved. It has been proved that the method which combine the genetic algorithms with BP Neural Networks to simulate the vehicle-loss model meet the requirement of the engineering project.