Fast Fingerprint Database Maintenance for Indoor Positioning Based on UGV SLAM
Fast Fingerprint Database Maintenance for Indoor Positioning Based on UGV SLAM
复制标题
基于UGV SLAM的室内定位指纹数据库快速维护
DOI：
10.3390/s150305311
复制
发表时间：
2015-03-01
期刊：
影响因子：
3.9
通讯作者：
Chen, Ruizhi
中科院分区：
文献类型：
--
作者：
Tang, Jian;Chen, Yuwei;Chen, Ruizhi
室内定位技术在过去二十年中变得越来越重要。利用接收信号强度指示器（RSSI）指纹的信号的opperity（SOP）是一种有前途的替代导航解决方案。然而，由于RSSI在操作期间由于其物理性质而变化并且容易受到环境变化的影响，因此室内指纹识别方法的一个挑战是以及时和有效的方式维护RSSI指纹数据库。基于自主研发的无人地面车辆（UGV）平台NAVIS，提出了一种指纹数据库快速更新的解决方案。NAVIS上安装了多个SOP传感器以收集室内指纹信息，包括收集磁场强度的数字罗盘、收集光强的光传感器以及收集预安装WiFi网络的接入点号码和RSSI的智能手机。NAVIS平台生成室内环境地图并在地图处理过程中收集SOP，然后对SOP指纹数据库进行真实的插值和更新。现场试验进行了评估所提出的方法的有效性和效率。实验结果表明，与传统方法相比，该方法能够以更高的采样频率（5Hz）和更密集的参考点快速生成和更新指纹数据库，并且无需先验信息即可生成室内地图。此外，指纹室内定位还可以快速检测环境变化。
Indoor positioning technology has become more and more important in the last two decades. Utilizing Received Signal Strength Indicator (RSSI) fingerprints of Signals of OPportunity (SOP) is a promising alternative navigation solution. However, as the RSSIs vary during operation due to their physical nature and are easily affected by the environmental change, one challenge of the indoor fingerprinting method is maintaining the RSSI fingerprint database in a timely and effective manner. In this paper, a solution for rapidly updating the fingerprint database is presented, based on a self-developed Unmanned Ground Vehicles (UGV) platform NAVIS. Several SOP sensors were installed on NAVIS for collecting indoor fingerprint information, including a digital compass collecting magnetic field intensity, a light sensor collecting light intensity, and a smartphone which collects the access point number and RSSIs of the pre-installed WiFi network. The NAVIS platform generates a map of the indoor environment and collects the SOPs during processing of the mapping, and then the SOP fingerprint database is interpolated and updated in real time. Field tests were carried out to evaluate the effectiveness and efficiency of the proposed method. The results showed that the fingerprint databases can be quickly created and updated with a higher sampling frequency (5Hz) and denser reference points compared with traditional methods, and the indoor map can be generated without prior information. Moreover, environmental changes could also be detected quickly for fingerprint indoor positioning.