Research And Implementation Of BDS/MEMS-SINS Loosely Integrated Sustainable Navigation Algorithm

With the rapid development of integrated navigation technology,satellite / inertial loosely integrated navigation system is widely used for its convenience in engineering.However,when the satellite signal is unlocked,the satellite / inertial loosely integrated navigation system will be transformed into pure inertial navigation system,which results in a sharp decrease in accuracy.In order to solve this problem,this paper studies the integrated sustainable navigation algorithm of BDS/MEMS-SINS and its embedded implementation method.First of all,the solution principle of strapdown inertial navigation system and the basic principle of BDS/MEMS-SINS integrated navigation system are studied.The BDS/MEMSSINS integrated navigation scheme based on estimation filtering and considering the heading measurement information is designed,and the state equation and measurement equation of the integrated navigation system are derived.The existing aerial attitude measurement hardware system of the research group is improved,and the dual antenna Beidou satellite navigation receiver is added.The hardware system of BDS/MEMS-SINS integrated navigation system is composed of stm32h743 and tms320c6748 as the processing core.Secondly,in view of the advantages of high accuracy and high speed of central difference Kalman filter(CDKF),a robust adaptive cdkf(RACDKF)algorithm based on innovation reconstruction and robust M-estimation is proposed.The algorithm uses innovation reconstruction method to solve the problem of short-term abnormal measurement information,improves the estimation accuracy and robustness of the algorithm,and uses robust M-estimation method to estimate the measurement noise in real time,so as to adapt to the sudden change of statistical characteristics of measurement noise in the filtering process.On this basis,the data fusion method of BDS/MEMS-SINS is designed when BDS navigation signal is normal.Thirdly,aiming at the problem that the precision of BDS/MEMS-SINS integrated navigation system drops sharply when BDS satellite navigation signal loses lock,a BDS/MEMS-SINS integrated navigation method based on long short term memory(LSTM)assisted by racdkf is proposed.When the Beidou satellite navigation signal is normal,the position error,velocity error and heading angle error estimated by RACDKF are used to correct the sins solution results,and the LSTM network for predicting the position error,velocity error and heading angle error is trained;when the Beidou satellite navigation signal is out of lock,the position error,velocity error and heading angle error predicted by LSTM network are used to correct the sins solution results.The experimental results show that this method can improve the accuracy and stability of the BDS/MEMS-SINS integrated navigation when the Beidou satellite navigation signal is out of lock.Fourthly,the BDS/MEMS-SINS integrated navigation method is embedded into the hardware system.The data communication logic of STM32H743 and TMS320C6748 is designed,the program of STM32H743 and TMS320C6748 are written,and the system performance is verified by sports car experiment.The experimental results show that the BDS/MEMS-SINS integrated navigation system can continuously output effective navigation information when the satellite signal is unlocked.The velocity error is not more than 0.42m/s,the latitude error is 3.35082e-05 degree(3.68m),the longitude error is1.0695e-05 degree(1.17m),the heading angle error is not more than 2.4 degree,the pitch angle error is not more than 2.3 degree,and the roll angle error is not more than 2 degree.The research results of this paper provide a reference for the sustainable design of BDS/MEMS-SINS loosely integrated navigation,and have great significance for promoting the development of Beidou navigation technology.