Multisensor track initiation method that addresses the missing measurement problem

A method for integrating multisensor data for the purpose of track initiation using horizon infrared and radar data is proposed. This multisensor track initiation (MSTI) method extends contemporary data fusion techniques so as to address the problem of missing measurements. The missing measurement phenomenon occurs due to a variety of reasons, the foremost of which is variation in sensor detection performance due to environmental factors.

The proposed MSTI method requires only the results of spatial feature tests that are performed on sensor data sequences. The formation of data sequences and the derivation of feature tests to integrate horizon radar and infrared data of differing resolutions is addressed.

Results are presented that detail the performance of the MSTI technique when operating on simulated data. It is shown that the statistical performance of the MSTI technique is better than or equal to that of the AND algorithm for a representative set of scenarios. The sensitivity of the MSTI method to variations in assumed feature test and data sequence statistics is also addressed.