A Maneuver Detection-Range Parameterized Cubature Kalman Filter for Bearings-Only Maneuvering Target Tracking

A maneuver detection-range parameterized cubature Kalman filter (MD-RPCKF) is proposed for bearings-only tracking of ground maneuvering target. Firstly, based on the range parameterization algorithm, the detection range of the observation station is divided into several sub intervals using the equal ratio principle, and each interval operates a cubature Kalman filter. Secondly, based on the measured likelihood function, the weights of the sub filters are updated and the target maneuver is detected. If the target is detected as maneuvering, new sub filters are generated to solve the problem of poor robustness and divergence of the filter, and the calculation amount is reduced by pruning and merging the sub filter. Finally, the state information of each sub filter is weighted and fused to estimate the target state accurately. The simulation results show that this algorithm can improve filtering stability and robustness, solving the problem of the decrease in tracking accuracy caused by unknown initial range and target maneuvering.