A Track Correlation Algorithm Based on Cross-covariance of Measurement Residuals

摘要

It is demonstrated in this paper that measurement residuals, of two sensor-level tracks from the same target are correlative because they have the common process noise, while residuals of two sensor-level tracks from different targets are irrelevant. A track correlation algorithm based on cross-covariance of residual is developed. The merit of this new developed algorithm is that its performance will not decrease in a high-density target environment. Simulation results are also given in this paper, which agree with theoretical analysis very well.