Selective Fusion of Out-of-Sequence Measurements with EK-IMM Estimator

In multisensor track fusion systems, according to the transmit or processing delay, the measurements can come to the fusion center out of sequence. In order to avoid either a delay in the output or the need for reordering and reprocessing an entire sequence of measurements, such measurements have to be processed as out-of-sequence measurements (OOSMs). Most of the algorithm dealing with OOSMs considered to always process these OOSMs. But sometimes dealing with OOSMs may not bring the improvement of track accuracy. Instead of always processing the OOSMs, we present a extended Kalman filter(EKF) based selective fusion method for the OOSM. Through assess the impact of OOSM data, determine the threshold and select the OOSMs to be fused. In order to make the algorithm suitable for hybrid stochastic dynamic system, we extension the selective fusion mechanism into the extended Kalman based interacting multiple model(EK-IMM) estimator. Simulation results are presented using measurements from Automatic Dependent Surveillance-Broadcast(ADS-B) and Secondary Surveillance Radar(SSR). lt is shown that selective fusion with EK-IMM estimator can reduce computational costs while maintaining near optimal performance.