Research On Technology Of Multi-sensor Fusion

Multi-sensor sensor provides a purposeful description of the environment that single sensor can not offer. Fusing several types information enhances the capability of intelligent system and yields more meaningful information otherwise unavailable or difficult to acquire by a single sensor modality. Multi-sensor fusion is an emerging technology related to information science, computer science and automatic science.In this dissertation, the composition, the existing problems and the future research directions of fusion systems are stated. We emphasize on the study of fusion architectures and fusion methods. The models of function, architecture and mathematics for multi-sensor systems are provided, which compose the core problem of fusion systems. The general approach to the design of fusion system is also discussed.Bayesian illation and D-S theory, which are used most in the fusion technology of parameter illation, are based on the statistic theory and implement the identity of targets in the observation space on the basis of prior information of the observation space. We discuss the theory of these two methods and validated their validity by simulation.We propose a method of fusing uncertain information Based on fuzzy integral, an intelligent fusion system is given, which can solve the FEI-DEO and DEI-DEO fusion problems. Last, we suggest a method based on the fusion theory and optimal weight. This method combines the advantages of these two methods, improves the fusion precision and enhances the capability of handling fuzzy information.We analysis the model and principle of multi-sensor fusion with two typical NN——BP network and RBF network. It first trains the NN with samples and then uses the trained NN as the state sort machine to do the object identity of multi-sensor fusion. This not only enhances the capacity of handling the uncertain information, but also quickens the fusion speed of system.