A new multi-sensor hierarchical data fusion algorithm based on unscented Kalman filter for the attitude observation of the wave glider

Abstract

In order to improve the accuracy of a single low-cost attitude observation sensor of the wave glider (WG), a new multi-sensor hierarchical data fusion algorithm is proposed. The bottom-level fusion strategy is the complementary data fusion of the accelerometer and the magnetometer, and the global optimal estimation of the bottom-level fusion result and the gyroscope based on the unscented Kalman filter (UKF) algorithm in the top-level fusion strategy. However, the estimation accuracy of the standard UKF algorithm depends on the accuracy of the noise model. Inaccurate estimation of the covariance matrix of the state noise and the measurement noise will cause the accuracy of the UKF algorithm to decrease or even diverge. By introducing the covariance matching technology, the innovation sequence and the residual sequence are used to realize the real-time estimation of the covariance matrix of state noise and measurement noise. At the same time, in view of the characteristic that the measurement equation is a strongly nonlinear equation, an improved covariance matching adaptive UKF algorithm is proposed. It effectively overcomes the limitations of the standard UKF and improves the adaptability of the multi-sensor hierarchical data fusion algorithm. Numerical simulation and experiment are carried out and the results demonstrate that the multi-sensor hierarchical data fusion algorithm is effective in solving the precision problem of the WG low-cost attitude observation unit.