A probabilistic nearest neighbor filter for m validated measurements.
Taek Lyul Song, Sang An Shin · 2003
The simplest approach for tracking a target in a cluttered environment is to select the validated measurement that is closest to the predicted measurement and use it in tracking as ifit were the true one. This method is the so-called neighborfilter(F) widely used for tracking in cluttex On the other hand, the probabilistic nearest neighborfilter(PNNF) is designed not to ignore the fact that the NN would be the false measurement and to calculate the probability of the event that the NN is iarget- originated. The PNNF algorithm does not utilize the current number of validated measurements that may be helpfir1 for calculating more reliable estimates in the realistic situation where the spatial density of false measurements is unknown. Incorporating the number of +,alidated measurements into design of the PNNF produces new data association proposed in this paper This filter has similar perjonnance to the PNNF if the spatial clutter density is known and it is less sensitive to the unknown spatial density offalse measurements.