非均匀场景下的协作定位基本限的研究

针对实际救援场景中由障碍物、复杂环境等因素导致的定位精度差的问题,提出将协作定位的思想应用于被困人员非均匀分布的实际场景中,在此基础上进一步提出了该场景下高效、低成本信标节点的布置.首先分析了协作定位系统的原理及其参数模型,并利用等效费歇尔信息矩阵(equivalent Fisher information matrix,EFIM)得到协作定位精度的基本限,即平方位置误差界(squared position error bound,SPEB),然后模拟2个实际中典型被困人员非均匀分布场景,在此基础上仿真,并比较模拟场景下协作模式与非协作模式的SPEB值,以及该场景下不同信标节点布置的SPEB值.仿真结果表明,被困人员之间的协作能够明显提高其自身定位精度,实际场景下信标节点的布置需要综合考虑建筑结构、成本、精度等因素.

Fundamental limit of cooperative positioning under non-uniform scene

To achieve high positioning accuracy in an actual rescue scene which is usually limited by obstacles and hash conditions, this paper proposes a cooperative localization system for a non-uniform scene in which the trapped people have the ability to communicate with each other. Basing on this point, we further put forward a low-cost and effective deployment of beacons.Firstly, the principle of a cooperative localization system as well as its parameter model is analyzed theoretically. Then we use equivalent fisher information matrix (EFIM) to calculate the squared position error bound (SPEB), which is the fundamental limit of localization accuracy. Finally, we simulate two typical actual scenes in which the distribution of trapped people is non-uniform. On this basis we simulate and compare SPEB in the cooperation situation with SPEB in non-cooperation situation. Additionally, the SPEBs which are calculated in different arrangements of beacons are also concerned. The results show that the cooperation among trapped people can improve positioning accuracy obviously and the deployment of beacons should take into account the structure of buildings, cost and positioning accuracy, etc.