Study On Dual-antenna GNSS Integrated Attitude Measurement

In the global satellite navigation systems,the attitude determination based on carrier phase technology has become a new research hotspot after the position determination and timing calibration.Compared with traditional attitude determination,attitude determination based on satellite navigation system has the advantages of low cost,high accuracy,strong robustness and no accumulated error,which makes it widely used in automatic driving,attitude control,mountain,bridge,dangerous building monitoring and other fields.The traditional inertial navigation attitude measurement system does not rely on external input information and is not susceptible to external interference,but the accumulated error will lead to a decrease in accuracy over time.Therefore,it is feasible to adopt the Integrated attitude measurement method to overcome the limitations of single satellite navigation system measurement and optimize the calculation process to improve the calculation rate.Therefore,this paper focuses on the following contents:(1)In this paper,the core problem of the satellite navigation system-the integer ambiguity resolution is studied in depth.From the effectiveness and timeliness of the ambiguity resolution,a Chaos Yield Sparrow Regression Search Algorithm(CY＿SRSA)is improved for the integer ambiguity search.Compared with Least-squares Ambiguity Decorrelation Adjustment(LAMBDA)algorithm,Modified Least-squares Ambiguity Decorrelation Adjustment(M-LAMBDA)algorithm,traditional Genetic algorithm(GA),Adaptive Genetic algorithm(AGA)and Sparrow algorithm(SSA),the performance of the algorithm is verified by real-time location solution experiment.(2)The dual-antenna dual-difference carrier phase difference equation is established,the relationship between attitude angle and baseline vector is introduced,and the fitness function model of attitude angle is constructed.An adaptive weighted particle swarm optimization algorithm(AWPSO)based on chi-square test is proposed.The algorithm improves the traditional PSO algorithm and utilizes the statistical characteristics of attitude search results,Chi-square test is used to eliminate the local optimal solution through the Mahalanobis distance of the attitude angle to ensure the global optimal result.The experimental results show that the static test error of the new algorithm is better than the direct solution algorithm,the least squares attitude estimation algorithm and the PSO-based attitude angle solution algorithm under the short baseline.(3)Based on the research of loose combination,the mathematical model of GNSSIMU’s Kalman filtering is deduced,the development of Kalman filtering and the existing problems are analyzed,and a constraint-based Attenuation Memory Cubature Kalman filtering algorithm(CAM＿CKF)is improved based on the volume Kalman filtering algorithm.The measurement noise matrix and covariance matrix are dynamically adjusted by weakening the innovation weight of the past moment to avoid the generation of filtering divergence in the filtering process and improve the filtering accuracy.Finally,through static and dynamic experiments under short baseline,the performance of Kalman filter,extended Kalman filter,Unscented Kalman filter and Cubature Kalman filter are compared and analyzed,and the effectiveness and reliability of the improved algorithm are proved.