Optimal design of the variable structure IMM tracking filters using genetic algorithms

摘要

Genetic algorithms for optimization of the multiple model and variable structure estimators are discussed in this paper. The estimation algorithm based on the multiple model and variable structure, are the best approach used in many systems, including maneuvering target tracking, noise recognition, etc. The RAMS algorithm, asserts that a multiple model algorithm consists of three steps: model set adaptation, initialization of model-based filters, and estimation. The first step, i.e., model set adaptation, is unique for VSMM algorithm and is the only superiority of the VSMM over FSMM. After the graph theory is used for this step and the sub-optimal switching digraph algorithm is discussed, we try to use the genetic algorithm for optimizing the thresholds used in the sub-optimal algorithm. The simulations show the improvement of the system performance when we use the optimal variable structure multiple model approach.