Study On Intelligent Navigation System

With the development of aeronautics, space and marine science, the performance of modern navigation systems cannot be satisfied by monotonous or traditional integrated navigation system. Intelligent Navigation Systems, which combined Artificial Intelligence and classical navigation technology, can achieve the self-management of multi-sensor devices. It also utilizes the data fusion theory, a newly developed cross discipline, to improve precision, reliability and intelligibility of navigation systems.Regarding the study of intelligent navigation techniques, the report focuses on the following aspects:Firstly, the report analyzes the composition and functions of intelligent navigation system, probes its combined patterns and finishes the overall design. The system adopts three-level comprehensive patterns. With the whole system software using message-driven based OOP method, the system consists of ten modules such as overall-command and ship movement. The conversion and transfer of navigation information, shift of patterns and data optimization are accomplished by intelligent navigation console.Secondly, Multi-sensor data fusion techniques play a crucial role in intelligent navigation systems. The report discusses the algorithm, structure and information allocation of federal filter, proposes the scheme and fusion algorithm of INS/GPS/Doppler Log based on multi-sensor federal Kalman filter. The simulation results show that the method is equivalent of centralized Kalman filter and that the former reduce the computation work and upgrade the reliability and fault-tolerance of the system.Thirdly, The report further studies the Intelligent Collision Prevention Expert System for Navigation (NICPES). By understanding and analyzing of navigation rules, and by the navigation experience and samples' collection and trim, the paper puts forth and establishes a multi-unit and layering KB systematic structure, and implement the KBM. According to features of different knowledge, we adoptmultifarious KR, such as: frame KR, production rule KR, procedure KR and neural network KR. We also build a multi-inference system, which based on analogy inference, forward illation inference, conversion inference, neural network inference and meta-rule inference. At the same time, we develop each reasoning algorithm. Otherwise, we also discuss multi-ship encounter situation in specially and build a mathematical model.Finally, the report discusses and establishes the theoretical framework of intelligent diagnosis. Incorporating fuzzy logic, neutral network and knowledge engineering, the report proposed theory and design of intelligent diagnosis system.