Data fusion of ALV GPS/DR integrated navigation system based on BP neural network

Abstract

Integrated navigation system of ALV is discussed in this paper, a data fusion method based on BP (Back Propagation) neural network is proposed for ALV's GPS/DR integrated navigation. System models have been established based on this data fusion method. Integrated navigation system uses GPS parameters as criterion to judge the validity of GPS. When GPS is valid, neural network is adopted for state estimation, which is four-layered network with 5-input/3-output neurons and two hidden layers. When GPS is invalid, DR is introduced by using outputs from IMU and odometer and initial information from BP network. Training and simulation are made with this BP network based integrated navigation system, whose performance is improved according to the simulation. The experimental results indicate that the navigation accuracy of integrated system has been improved in comparison with that of either GPS or DR.