A fuzzy-neural networks approach for multisensor fusion

There are limit in the multisensor fusion with the conventional fusion methods. Artificial intelligence fusion methods, such as fuzzy logic and neural networks, have received more attention in the recent years. By combining the advantages of fuzzy logic and neural network theory, a fuzzy neural network approach is proposed. Using fuzzy reference describes the states of sensors can avoid sensors' faults or failures. Then reasoning results are taken as inputs of the neural network. The approach can decrease the number of parameters to be trained and increase training efficiency and reduce the complexity. This paper presents the architecture of multisensor fusion based on Fuzzy-Neural network and describes and discusses data fusion algorithm in detail.