The application of BP neutral network optimized by genetic algorithm in transportation data fusion
Xiaomei Sun, Zhaosheng Yang, Zhijian Wang · 2010
BP neutral network is a method to fuse datas from several dissimilar sensors for more precise results, but its low convergence rate and easily falling into local minimum often decrease the fusion precision. In order to overcome this drawback, we combined BP neutral network with genetic algorithm, and optimized connection weights and threshold value of BP neutral network by genetic algorithm. Then the optimal BP neutral network was applied in transportation system for traffic volume fusion. Experiment result demonstrates that our method satisfied the real-time request of transportation system more effectively.