Research And Implementation Of High-Precision Fusion Direction Finding Algorithm Based On IMU/GNSS

With the wide use of direction finding technology in the field of navigation and the Internet of things,the high-precision direction finding technology based on low-cost sensors and short baseline appears to be particularly important.At present,this technology has some obvious shortcomings.On the one hand,the inertial sensor(IMU)-based direction finding algorithm has accumulated errors,which cannot provide reliable heading information for a long time.On the other hand,the direction finding algorithm based on the Global Navigation Satellite System(GNSS)is greatly affected by GNSS signals and it is difficult to obtain high accuracy and continuous heading information in an scenes with poor signal quality.Therefore,in order to improve the accuracy and real-time performance of direction finding based on low-cost sensors and short baseline,a set of fusion direction finding algorithm based on IMU/GNSS is proposed in this thesis.First of all,we propose an improved dual GNSS antenna direction finding algorithm aiming at the problems of low accuracy and poor real-time performance.The improvement includes the following three aspects.(1)To reduce the search space of the fixed ambiguity algorithm based on the known baseline length so as to improve the success rate of solution with the dual antenna GNSS direction finding.(2)Adopting the multi-antenna design,two dual antenna GNSS direction finding subsystems are introduced to improve the accuracy and real-time performance of GNSS direction finding.(3)A verification algorithm based on baseline length and verification value is designed to ensure the reliability of the solution.Secondly,this thesis designs a combined direction finding algorithm based on the Extended Kalman Filter(EKF),which guarantees that it can still provide reliable heading information under the condition of poor GNSS signals and has good real-time performance.Meanwhile,for purpose of adapting to direction finding in various scenarios,this thesis proposes a dynamic parameter adjustment algorithm based on the carrier-to-noise ratio,position accuracy strength(PDOP)and satellite elevation angle of the GNSS signal.Finally,based on the algorithm proposed in this thesis,a direction finding system based on low-cost sensor and short baseline is designed and implemented.The hardware design of the system is based on the inertial sensor MPU9150 and GNSS receiving module NEO-M8T.The software of the system is based on the API provided by the open source software RTKLIB to implement the algorithm proposed in this thesis.And a lot of tests are carried out in the actual environment.The test results show that the RMSE of the algorithm is 0.45 and the average frequency is 326 times per minute.It is proved that the algorithm can obtain high precision and real-time heading information under the condition of low-cost sensor and short baseline.