中文摘要：

针对卫星信号到达地面时较弱、不能穿透建筑物的问题,该文对无法使用卫星定位时的辅助手段之一——室内定位技术作了研究。室内定位多传感器信息融合需要在统一坐标框架下进行,传统的空间配准解决方案多为简单的“松组合”加工。该文给出了多传感器空间配准数学模型,通过构造极大似然函数估计传感器间的位置偏差,采用粗估计与高斯牛顿迭代估计的两步法求解,并用仿真算例验证了该文所采用算法的有效性与正确性。

英文摘要：

As an auxiliary positioning method when satellite positioning cannot be used, indoor positioning technology can effectively solve the problem that the satellite signal arriving at the earth is too weak to penetration buildings. Multisensor information fusion for indoor positioning requires a canonical coordinate frame to accommodate the outputs from different sensors. The conventional spatial registration methods are mainly loosely coupled processing methods. This paper showed the mathematical mode for multisensor spatial registration and estimated the location registration by constructing the maximum likelihood function, and the mathematic models were solved using two-step methods, including the coarse estimation and Gauss-Newton iterative estimation. The effectiveness and correctness of the proposed method were verified through a numerical simulation example.