节点文献

基于惯性传感器和楼层平面图的室内定位

Indoor positioning based on inertial sensor and floor plan

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 陈建欧钢彭敖郑灵翔石江宏

【Author】 Jian Chen;Gang Ou;Ao Peng;Lingxiang Zheng;Jianghong Shi;School of Information Science and Engineering,Xiamen University;

【机构】 厦门大学

【摘要】 由于智能手机扫描Wi-Fi信号消耗过多的电量,因此基于智能手机进行长时间室内定位提出了挑战。本文提出了一种结合惯性传感器和楼层平面图的混合定位方法。惯性定位系统积分加速度和角速度获得行人的相对位移。由于惯性传感器的偏差、非正交性以及用户抖动,惯性定位系统的主要问题是方向漂移。惯性定位误差随着时间进行没有边界的增长。考虑到用户不能越过墙壁等障碍物,我们提出了基于楼层平面图的粒子滤波来修正惯性定位系统的方向漂移问题。结合楼层平面图,所有的粒子被分为两类:没有越墙粒子和越墙粒子;如果粒子没有越过墙壁,粒子按照高权重参与重采样;当粒子越过墙壁时,降低粒子参与重采样的权重。为了减少越墙粒子,我们使用萤火虫算法修改越过墙壁的粒子。萤火虫操作修改越墙粒子的位置让越墙粒子变成没有越墙粒子。在萤火虫算法之后,采用一种自适应算法更新粒子权重。因此,本文提出了萤火虫粒子滤波来提高惯性传感器和楼层平面图的室内定位精度。在打电话和手持智能手机两种姿态下,定位结果显示走廊和自习室平均定位精度分别为2.24米和1.47米。

【Abstract】 Since smartphones scan Wi-Fi signals to consume too much electricity,it is challenging to use smartphones for long time indoor positioning.This paper proposes a hybrid localization method combined with inertial sensors and floor plan.The relative displacement of pedestrian is obtained by integrating acceleration and angular velocity of an inertial positioning system.Due to the sensor deviation,non-orthogonality and pedestrian jitter,the main problem of inertial positioning system is heading drift.An inertial positioning error has no boundary growth over time.Considering that a user cannot cross the wall and other obstacles,we propose a particle filter based on a floor plan to correct heading drift.Combined with a floor plan,all particles are divided into two categories:wall-crossing particles and non-wall-crossing particles;if particles did not cross the wall,they involved in the resampling according to the high possibility;if particles crossed over the wall,they were reduced their weight in resampling.In order to further reduce the wall-crossing particles,we use firefly algorithm to modify the wall-crossing particles.The wall-crossing particles become the non-wall-crossing particles.In addition,an adaptive algorithm is used to update the modified particle weights.Therefore,a firefly particle filter was used to improve the indoor positioning accuracy.The positioning results show average positioning accuracy of corridor and study room are 2.24 and 1.47 meters,respectively under two postures of calling and handheld.

  • 【会议录名称】 第九届中国卫星导航学术年会论文集——S10 多源融合导航技术
  • 【会议名称】第九届中国卫星导航学术年会
  • 【会议时间】2018-05-23
  • 【会议地点】中国黑龙江哈尔滨
  • 【分类号】TP212.9
  • 【主办单位】中国卫星导航系统管理办公室学术交流中心
节点文献中: 

本文链接的文献网络图示:

本文的引文网络