An Algorithm for Tracking Multiple Targets-英文文献.pdf

IEEII TRANSAC1~1ONSON AUTOMATIC CONTROL, VOL. Ac-24, NO. 6, DECEMBER 1979 843 An Algorithm for Tracking Multiple Targets DONALD B. REID, MEMBER, IEEE Abstract—An algorithm for tracking multiple targets In a cluttered algorithms. Clustering is the process of dividing the entire environment Is developed. The algorithm Is capable of Initiating tracks, set of targets and measurements into independent groups accounting for false or m[~clngreports, and groom sets of dSPeUIIt ~or clusters~ Instead of solvin one lar e roblem a reports~As each measurement Is received, probabilities are ~*uI~ted for ~ g g ~‘ the hypotheses that the measurement ~ f~ ~ IUM)Wfl ~ number of smaller problems are solved in parallel. Fi- in a target file, or from a new target, or that the measurement Is false. nally, it is desirable for an algorithm to be recursive so Target states are estimated from each such da*a-as.soclatloo hypothesis that all the previous data do not have to be reprocessed using a 1C~InlQnfilter. As mere measurements are received, the probabill- whenever a new data set is received. ties of Joint hypotheses are calculated recursively using all available The a! onthm can easurements fro inforv~~Hnsuch as density of wiknown targets, density of false ~ g use m rn two ~.~er- probability of ietectlon, ami location ~rtainty. mis ~iciiing tecii- cut generic types of sensors. The first type is capable of nique allows correlation of a asurement with Its a based on sending in