A combination algorithm of Chaos optimization and genetic algorithm and its application in maneuvering multiple targets data association

The most important problem in targets tracking is data association which may be represented as a sort of constraint combinational optimization problem. Chaos optimization and adaptive genetic algorithm were used to deal with the problem of multi-targets data association separately. Based on the analysis of the limitation of chaos optimization and genetic algorithm, a new chaos genetic optimization combination algorithm was presented. This new algorithm first applied the rough search of chaos optimization to initialize the population of GA, then optimized the population by real-coded adaptive GA. In this way, GA can not only jump out of the trap of local optimal results easily but also increase the rate of convergence. And the new method can also avoid the complexity and time-consumed limitation of conventional way. The simulation results show that the combination algorithm can obtain higher correct association percent and the effect of association is obviously superior to chaos optimization or genetic algorithm separately. This method has better convergence property as well as time property than the conventional ones.