区域生长辅助的地图配准在室内定位中的应用

地图匹配（ MM）算法通过粒子滤波（ PF）利用室内地图信息来抑制基于惯性传感器的室内定位系统的误差累计。利用区域生长（ RG）算法结合当前步长和方向信息在地图上找到合理的落脚范围，并以此来判断粒子的有效性。这种方法能有效改善地图配准算法的实用性和计算复杂度。提出一种改进的零速度（ ZV）检测算法能准确提取步伐信息，间接提升了零速度更新（ ZUPT）算法和地图配准算法的精度。实验结果表明：该算法的定位误差小于1.0%，定位精度比单纯的航位推算（ DR）算法平均提高了5.97%。

Application of map-matching aided by region growing in indoor positioning

Map-matching( MM )algorithm is used to inhibit accumulation of error of indoor positioning system based on inertial sensor with particle filtering( PF)and indoor map. Use region growing( RG)algorithm,combined with step length and orientation,find reasonable area on indoor map,and decide validity of particles according to this. This method can reduce computational complexity and improve practicality of MM algorithm. An improved zero velocity( ZV)detection algorithm is proposed,which can extract step information accurately,and indirectly enhance precision of zero velocity update( ZUPT )algorithm and MM algorithm. Experimental result shows that localization error of the proposed algorithm is less than 1. 0% and the precision is increased by 5. 97% on average compared with dead-reckoning( DR)algorithm.