Optimal multiesensor asynchronous fusion based on state transform

Asynchronous fusion is one of main difficulties in multisensor data fusion field at present. Aiming at the problems which exist in current asynchronous fusion methods based on pseudo-measurements, such as man-made noise correlations, bad real-time tracking performance, and excessive computation burden at the final fusion time etc., this paper proposes an optimal sequential asynchronous fusion algorithm based on state transform as a result of researching continuous and distributed multisensor asynchronous dynamic systems. The proposed algorithm establishes dynamic state equation between two adjoining sample times by use of state relation obtained from the discretization process of continuous systems between these sensor sample times during fusion period and systemic fusion time. Accordingly, it can obtain a series of new state and measurement equations which accord with the requirements of standard Kalman filter in the fusion period, and asynchronous data fusion is finished by using these measurement and new state equations to perform sequential Kalman filter. The theoretic performance analysis and computer simulation show that the proposed method can not only avoid and overcome these current problems but also can achieve high fusion estimate accuracy.