The Research Of Nonlinear Filtering Method In WSN/INS/Vision Indoor Integrated Navigation

The indoor navigation system consist of wireless sensor networks (WSN), inertial navigation systems (INS) and vision navigation combines the advantages of the three above, with the characteristics of low cost, high precision and good reliability, is not only an ideal combination, but also the development trend of small range of high-precision navigation in the future. WSN/INS/Vision integrated navigation system typically has two kinds of combination methods, loose coupling integrated navigation and tight coupling integrated navigation. Between them, tight coupling uses the original measurement information, with no correlation among the measurement input, and has higher precision for navigation than loose coupling. The author of this paper make a systematic research on the nonlinear filtering method in tight coupling integrated navigation system and make a series of simulation experiments.This paper focuses on the combination of three navigation model, combining the nonlinear filter, and carries out the following studies:Firstly, describe the development of navigation, then introduce the basic principles and methods of various navigation methods, focus on the relevant principles of SINS, ranging and positioning theories of WSN, vision navigation method based on the optical flow pyramid, etc.. Filtering in the navigation system is also described.Secondly, design the specifically combination of WSN/INS/Vision navigation system based on characteristics of each kind of navigation, then establish the integrated navigation model. Focus on the principles and characteristics of the combination of tight coupling integrated navigation system, establish the state equation and measurement equation for navigation system in connection with the characteristics of extended Kalman filter (EKF) and iterated extended Kalman filter (IEKF).Finally, design of the WSN/INS/Vision navigation simulation system based on the tight coupling combination. Make simulation experiments for each navigation systems, combine the simulation output data of gyroscopes and accelerometers, WSN positioning results and the measured date of visual navigation, then on the basis of the data above, design the nonlinear filters such as extended Kalman filter and iterated extended Kalman filter, and make a series of experimental studies. Experimental results show that IEKF method is more suitable for tight coupling combination for integrated navigation system compares with the traditional EKF method, the indoor integrated navigation system will make higher precision and better performance overall.