Multi-sensor Track-to-Track Fusion Using Simplified Maximum Likelihood Estimator for Maneuvering Target Tracking

The focus of this paper is to present the distributed architecture of track-to-track fusion for computing the fused estimate from multiple filters tracking a maneuvering target with the simplified maximum likelihood estimator. The architecture consists of sensor-based Kalman filters, local processors and global fuser. Each sensor tracker utilized in the reference Cartesian coordinate system is described for target tracking when the radar measures range, bearing and elevation angle in the spherical coordinate system. The Bar-Shalom track-to-track fusion algorithm is used in each local processor to merge two tracks representing the same target. The decoupled process is adopted to simplify the batch form of the maximum likelihood estimator due to the block-diagonal covariance matrix. The resulting global fuser can be implemented in a parallel structure to facilitate estimation fusion calculation. Simulation results show that the proposed fusion estimator has computational advantages over the maximum likelihood estimator with similar performance.