Leg-by-leg机动条件下纯方位跟踪的观测站机动策略

针对观测站机动前目标距离未知导致无法计算Fisher信息矩阵(Fisher information matrix, FIM)的问题, 从粗估目标初始距离和减小粗估误差对FIM行列式影响两方面着手, 提出在转向前计算FIM进而设计机动策略的方法。首先, 在修正极坐标系下建立系统模型, 描述了观测站leg-by-leg机动模式下FIM的计算方法。其次, 通过假设目标速度实现目标初始状态粗估, 提出使归一化初始距离粗估误差最小的目标速度假设方法。然后, 对FIM行列式进行近似及多项式展开, 通过控制前后leg段比例以减小状态粗估误差对FIM行列式的影响。最后, 利用仿真验证了所提机动策略设计方法的可行性和有效性。结果表明, 对于典型的水下目标跟踪场景, FIM行列式的有效估计概率为76.7%, 平均的相对估计误差为0.12。

Observer maneuver strategy for leg-by-leg bearing-only tracking

Aiming at the problem that the Fisher information matrix (FIM) cannot be calculated due to the unknown target distance before the observation station maneuvers, this paper proposes a method to calculate the FIM and then design the maneuver strategy from two aspects of roughly estimating the target initial distance and reducing the influence of rough estimation error on the determinant of FIM. Firstly, the system model is established in the modified polar coordinate system, and the calculation method of FIM in the leg-by-leg maneuver mode of observation station is described. Secondly, the initial state of the target is roughly estimated by assuming the target speed, and the target speed assumption method is proposed to minimize the error of the normalized initial distance rough estimation. Then, the determinant of FIM is approximated and expanded by polynomials, and the influence of state estimation error on the determinant of FIM is reduced by controlling the ratio of front and back leg segments. Finally, the feasibility and effectiveness of the proposed design method of maneuver strategy are verified by simulation. The results show that for the typical underwater target tracking scene, the effective estimation probability of FIM determinant is 76.7%, and the average relative estimation error is 0.12.