Location Research Based On MEMS-INS And GPS

In recent years,the emergence of autonomous car and other similar applications,the positioning accuracy and reliability of the navigation put forward higher requirements.On the one hand,low-cost inertial navigation system(INS)cumulative error is large,so the accuracy is still relatively poor;the other hand,GPS system due to delay and occlusion,its reliability is low.Combine them,making the higher reliability and accuracy of the navigation and positioning possible.The study includes a variety of error models,for GPS to fully consider the satellite clock error,receive clock error,ionospheric delay,tropospheric delay,multi-path error,satellite orbit error,measurement noise,and user equivalent distance error,speed and attitude equation,the stochastic process model is selected reasonably,the state equation of up to 29 state components is established,and then the linear and extended Kalman filter are used to complete the low cost INS / GPS fusion Positioning.The data of the GPS data were tested and the data fusion experiment was carried out with MATLAB software,and the positioning results were compared with the reference trajectory.The results show that GPS and INS are loosely coupled to each other and are suitable for applications with good satellite visibility.In applications where GPS interruption is frequent,the use of tight coupling is more reliable.Both solutions are more reliable and accurate than GPS or INS works alone better.