论文部分内容阅读
介绍了用于机动目标跟踪的变结构多模(VSMM)估计中可能模型集(LMS)算法。阐明了该算法通过剔除不可能模型,保留一般模型,激活与重要模型相邻的模型实现模型集的自适应原理,并给出了其实现步骤。采用LMS算法对恒加速(CA)机动目标跟踪的仿真结果表明,LMS算法的滤波效果优于交互式多模型(IMM)算法,算法简单易行、计算量小,可用于跟踪运动规律较复杂的目标。
The possible model set (LMS) algorithm in variable structure multimode (VSMM) estimation for maneuvering target tracking is introduced. It shows that the algorithm can eliminate the impossible model, keep the general model, activate the adaptive principle of the model set which is adjacent to the important model and give the realization steps. The simulation results of LMS algorithm for constant-speed (CA) maneuvering target tracking show that LMS algorithm has better filtering performance than IMM algorithm. The algorithm is simple and easy to implement and has a small amount of computation. It can be used to track the motion law more complexly aims.