经典的交互式多模型(IMM)算法由于转移概率矩阵固定不变,在跟踪运动状态未知且多变的临近空间高超声速目标时存在一定的不合理性。通过采用准贝叶斯(Quasi—Bayesian)估计对转移概率矩阵在线调整的方法,与基于CV—CA-Singer的经典交互式多模型算法进行融合形成了IMM—QB算法。Monte Carlo仿真实验表明,IMM—QB算法对临近空间高超声速滑跃式机动目标的跟踪精度有一定程度的提高。
Due to the fixed transition probability matrix (TPM), the classic interacting multiple model algorithm(IMM) is not suitable for directly use for tracking near space hypersonic strong maneuvering target, whose movement may vary as much as possible. By using the quasi-Bayesian estimate online adjusting method to adjust the TPM, an improved IMM algorithm based on CV-CA-Singer using such method is proposed. Results from Monte Carlo simulation experiments show that the new IMM algorithm is better than the classic algorithm when tracking near space hypersonic slippage leap maneuvering target.