In this thesis, we suggest an algorithm for multiple target tracking in the large missed-detections environment.
In the environment which has the physical handicap or clutter, the sensor can not provide enough detection for target tracking. In the large missed-detections environment, It is difficult to use previous study on multi-target tracking, i.e JPDA Filter or MHF. It does too frequently rise incorrect association.
We allow to many-to-many association between tracks and measurements. We reduce the redundant tracks by track-merge. This procedure will associate more exactly.
The proposed algorithm can be applied to the base function in the multi-sensor multi-target tracking system. And the extension of the proposed algorithm helps to track objects in noisy environment.