论文部分内容阅读
针对在未知但有界噪声假设下的双基阵纯方位目标跟踪问题,本文提出了一种基于外定界椭球的集员估计(EOB-SME)跟踪算法。该算法具有类似于Kalman滤波的预测-校正递推更新结构,并且在时间更新和量测更新递推阶段分别有一个加权参数。通过最小化估计误差的Lyapunov函数的上界来求取量测更新递推阶段的加权参数,减少了算法的计算量;同时将非线性系统线性化后所产生的误差用椭球进行外包,与量测噪声椭球组成新的噪声椭球。仿真结果表明:在有界噪声假设下,本文所提出算法对纯方位机动目标的跟踪精度更高。
In this paper, aiming at the target tracking problem of dual-array purely azimuth based on the unknown but bounded noise assumption, this paper proposes a novel EBS-C tracking algorithm based on outer bounding ellipsoid. The algorithm has a prediction-correction recursive update structure similar to Kalman filtering, and has a weighted parameter respectively in the time update and measurement update recursion phases. The upper bound of the Lyapunov function which minimizes the estimation error is used to obtain the weighted parameters in the recursion stage of measurement updating, which reduces the computational complexity of the algorithm. At the same time, the error generated after linearization of the nonlinear system is outsourced by ellipsoid, and Measure the noise ellipsoid to form a new noise ellipsoid. The simulation results show that the algorithm proposed in this paper has a higher tracking accuracy for purely azimuth maneuvering targets under the assumption of bounded noise.