摘要
设计了一种基于模糊逻辑推理的机动目标多模型跟踪新算法(FMMTA),把量测新息对其协方差的逆的加权二次函数作为模糊推理系统的输入,并通过模糊逻辑推理得到模型集中各模型的匹配度,代替了交互式多模型(IMM)算法中的模式概率计算,降低了计算的复杂度.该算法将测量空间的不确定性映射到模糊空间,从而解决了从测量空间的不确定性到模式空间不确定性的模糊推理问题,并将模糊推理与多模型卡尔曼滤波结合,进行并行处理,有利于机动目标的实时跟踪.MonteCarlo仿真结果表明,在模糊规则设计恰当的情况下,FMMTA算法相对于IMM算法在降低机动目标位置和速度的跟踪误差方面更有效.
A new tracking algorithm based on fuzzy logic inference for manoeuvring target, called the fuzzy multiple model tracking algorithm (FMMTA), is presented. In the new algorithm, the quadratic function of the filtering measurement innovation weighted with the inverse of its covariance matrix is considered as the input of the fuzzy inference mechanism to get the matched degree for each filtering model in the designed model set, by which the model probability in the existing interactive multiple model algorithm (IMM) is once replaced, and the calculation complexity is decreased obviously. The uncertainty of measurement space is mapped into the fuzzy space, and the uncertainty of measurement space is resolved via the fuzzy inference mechanism. Then the fuzzy inference and Kalman filtering are combined to run in parallel, and the result is very effective for tracking manoeuvring target in real time. The MonteCarlo simulation results indicate that it is more efficient to decrease the errors of the position and speed for manoeuvring target tracking by using the FMMTA rather than the IMM algorithm when the fuzzy rules are designed properly.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2002年第12期1240-1244,共5页
Journal of Xi'an Jiaotong University
基金
国家重点基础研究发展规划“九七三”资助项目(2001CB309404)
