中文摘要：

利用多个体之间的相对信息提高个体的定位精度的协同导航是一个很值得探讨的问题。自治水下航行器（AUV）在移动长基线（MLBL）网络中获取各应答器的位置信息并解算出相对各应答器的距离量测后,再通过贝叶斯滤波算法集中式提高其定位精度。但是,集中式利用量测的方法没有考虑由于水下环境的宽广性和水下传感器的限制导致的声信号在AUV和应答器间的往返时间内AUV的移动和应答器间位置的差异所带来的影响。首先以并行滤波的形式总结了集中式扩展卡尔曼滤波（EKF）协同导航方法,并针对集中式利用量测的不利因素,提出了按照异时量测的产生顺序即时更新AUV状态的更有效的序贯EKF协同导航算法,最后在仿真中将两种处理方法进行了比较。

英文摘要：

Cooperative navigation which can improve the individual localization accuracy through utilizing the relative information among multiple individuals is a problem worthy of discussing.An Autonomous Underwater Vehicle（AUV） can acquire the location information of the responders through depending on the Moving Long BaseLine（MLBL） network and calculate the relative distances between the responders and the AUV,and then improve its localization accuracy synchronously through a Bayesian filter.However,the method of the centralized use of measurements does not consider the influences brought by the movement of the AUV within the round-trip time of the acoustic signals spreading between the AUV and the responders and the location differences of the responders caused by the breadth of the underwater environment and the limitation of underwater sensor.In this paper,a cooperative navigation algorithm based on the centralized Extended Kalman Filter（EKF） is summarized in the form of a parallel filter firstly,and then a more efficient cooperative navigation algorithm that updates the AUV state immediately according to the producing order of asynchronous measurements based on the sequential EKF is proposed considering the adverse factors of the centralized use of measurements,and two processing algorithms are compared in the simulation lastly.