WGS-84坐标系下多空基无源传感器最大似然配准

Maximum likelihood registration for passive sensors of multiple airborne platforms in WGS-84

相比多运动平台有源传感器配准或异质传感器配准问题,多平台无源传感器的配准由于无距离信息将更为复杂,鲜有相关研究。为此,首先构建了WGS-84坐标系下有偏无源观测模型,然后将最大似然配准(maximum likelihood registration,MLR)算法扩展到空基多运动平台无源传感器的配准。运用复合函数求导链式法则,推导出应用MLR算法时至为关键的传感器观测量对目标状态的雅克比矩阵。为计算该矩阵,研究了WGS-84坐标系下两平台利用仅角度观测对目标的无源定位问题。理论和仿真结果表明该方法可实现无源传感器配准,配准误差逼近其Cramer-Rao界,验证了该方法的有效性。

Compared with those registration problems of active sensors or dissimilar sensors on multiple moving airborne platforms, the registration problem for passive sensors on different airborne platforms will be- come more complex owing to missing range measurement, and there is little relative literature. Thus, firstly, the biased measurement model for passive sensors based on world geodetic system-84 （WGS 84） coordinate sys- tem is constructed, and then the maximum likelihood registration （MLR） algorithm is extended to passive sen- sor registration for multiple moving airborne platforms in WGS-84 coordinate system. Using the chain derivative rule of composite function, when MLR is applied the key Jacobi matrix of sensor measurements to target state is derived. In order to compute the matrix, passive location problem for a target in WGS-84 by two airborne plat forms with angle-only measurements is investigated. Theory analysis and simulation results show that the meth- od can realize passive sensor registration, and the registration errors can approach the Cramer-Rao low bound, those indicate the validation of the algorithm.