Multisensor track association in the presence of bias

A central problem in multitarget-multisensor tracking is track-to-track association and fusion for tracks from multiple sensors. This problem is often confounded by the presence of inherent sensor biases, missing tracks, and false tracks. In this paper, an algorithm that addresses track-to-track association in the presence of bias and the corresponding bias estimation is presented and some parametric performance results are given. The algorithm utilizes Murty´sK-best algorithm to efficiently achieve a maximum likelihood estimate of the bias in conjunction with the most probable hypothesis for track-to-track association. Numerical examples are given to illustrate the application of the algorithm.