Study On The Algorithm Of Compact Combination Of Combined Navigation In Complex Environment

With the application and popularization of navigation technology,people's production and lifestyle are more dependent on the services of the navigation system,and higher requirements are placed on the accuracy and comprehensiveness of the navigation system.Tightly-coupled system is a combined navigation system that can deeply utilize the information of inertial navigation and satellite navigation.Due to the tightly-coupled system has high requirements on the accuracy of the model and has strong nonlinearity in practical applications.This paper studies the tightly-coupled system from the perspective of data processing,In-depth analysis of the tightly-coupled structure,analysis and improvement of the Tightly-coupled system filter algorithm,the main work and innovations of this article are as follows:1?This paper first introduces the working principle of BDS satellite system and inertial navigation system,analyzes the sources of error in integrated navigation and positioning,derives the state equation and measurement equation of the tightly-coupled system,and establishes an integrated navigation system error model.By designing the trajectory of the carrier,simulating the relevant data to complete the combined navigation solution and making a simple analysis,it provides an experimental platform for the simulation comparison of the following algorithms.2?Aiming at the problem that the adaptive extended Kalman filter's error covariance is easy to be non-definite in iteration,this paper improves the estimation method of system noise covariance.This algorithm reduces the computational complexity of the covariance matrix.At the same time,it adopts the method of predicting residuals to obtain the forgetting factor in real time,and applies it to the system noise covariance matrix,enhance the system's ability to suppress filtering divergence in the case of interference.3?In order to solve the complex calculation problem of traditional strong tracking unscented Kalman filter,this paper improves the traditional strong tracking unscented Kalman algorithm.First,make a specific analysis of the filtering period of traditional strong tracking unscented Kalman,derive and improve the solution of the fading factor,simplify the filtering process of unscented Kalman and strong tracking,and according to the degree of uncertainty of the different states of the system The size is adjusted in a targeted manner to construct a fading matrix to improve the filtering accuracy.