A Multi-Sensor Data Association Algorithm Based on Time Constraint

For the high real-time demands of multi-sensor target tracking system, a global optimum data association algorithm based on time constraint is presented. Firstly, the statistical distance between measurement and prediction center of the track is calculated. An optimal association model among all measurements and all tracks is founded. Then, an improved auction algorithm is proposed to solve the association model. Multi-time auction is conducted in the constrained time, and the optimal solution is obtained by comparing the value of objective function. Simulation shows that the proposed method has a perfect association performance in the constrained time.