Research On Integrated Navigation Method Based On Nonlincar Filter

The indoor and outdoor environment is becoming more and more complex,and the demand for navigation is increasing.Especially in large public places such as shopping malls,libraries,subway stations,etc.,it is often necessary to quickly obtain the location of the user.WIFI,infrared,UWB and other indoor positioning methods based on wireless radio frequency signals cannot be used in power outages and other situations.Outdoor positioning mainly uses mature GPS,but GPS.signals have problems such as occlusion.Strapdown inertial navigation(SINS)only needs to be based on the measurement data of the inertial measurement unit(IMU)to achieve independent and autonomous navigation.Therefore,GPS and SINS are often combined to achieve continuous indoor and outdoor pedestrian navigation.Achieving high precision and good real-time GPS/SINS integrated navigation requires a good filtering algorithm,so this paper mainly studies the nonlinear filtering method.Firstly,the commonly used navigation coordinate system and its transformation,GPS navigation principle,INS principle,GPS/INS combined navigation model are analyzed;the principle analysis and simulation comparison of EKF,UKF,PF under the Bayesian nonlinear filtering framework.Secondly,in view of the influence of colored noise caused by IMU,which leads to low navigation and positioning accuracy,modeling is performed according to the characteristics of IMU noise and the extended Kalman filter model is improved,and the improved algorithm is verified in integrated navigation.Finally,the principle analysis of the box particle filter(BPF)is carried out,and the PF and BPF are compared and simulated.It is found that BPF is better than PF in accuracy and real-time,especially real-time;establish a GPS/INS loose combination model,filter the box particle and The particles are used as combined filters for filtering,and the independently designed GPS/IMU/magnetometer information collection system is used for data collection.The experimental results show that the positioning accuracy and real-time particle filter have advantages.