基于Wi-Fi指纹与视觉融合的室内交通定位
|
摘要
为解决室内交通在GPS盲区情况下的定位问题,将Wi-Fi指纹定位与视觉定位相结合,研究了一种高精度室内交通定位方法。在离线阶段,在Wi-Fi采样点处采集无线AP的MAC地址及其采样点坐标,生成Wi-Fi位置指纹数据库,然后采集定位区域内的所有门牌图片,生成训练图像集,计算出训练图像的SURF与ORB全局特征描述符,并与标志采样点坐标共同构成视觉定位数据库。在定位阶段,采集待定位点的实测指纹及被测试图像,利用指纹匹配算法得到待定位区域的范围及坐标,再利用图像特征匹配算法与KNN算法在训练图像集中得到与被测试图像相匹配的训练图像,即匹配图像,通过查询视觉定位数据库得到视觉定位范围及坐标。当匹配图像的坐标位于Wi-Fi定位范围内,将视觉定位坐标作为最终的定位参考坐标,反之将Wi-Fi定位坐标作为最终的定位参考坐标。实验选取了不同的室内交通环境对算法进行验证,定位误差为0m的占比为82%,误差在6m之内的占比为12%,误差在6～10m之内的占比为6%;平均定位误差为0.75m,而且在线平均定位耗时仅为0.56s,能实现鲁棒的高精度的室内交通定位需求。
In order to solve problems of indoor traffic positioning in GPS blind area,a high-precision method is proposed by combining Wi-Fi fingerprint localization with visual localization.In offline stage,Wireless AP MAC address and sample point coordinates are collected to establish a database of Wi-Fi location fingerprint,and gather plate images to develop training image sets,then calculate the SURF and ORB global feature descriptor of the training images,a visual positioning database is developed with the sample point coordinates.In positioning stage,undetermined measured fingerprint images and tested images are collected.A matching algorithm for fingerprints is used to get regional scopes and coordinates,and a matching algorithm for image features is used to obtain training or matching images with KNN algorithm,by querying orientation of visual database for visual range and coordinates.When the coordinates of the matching images are located within the range of Wi-Fi location,the visual location coordinates will be used as reference coordinates for final location,or vice versa.Different indoor traffic environments are applied to verify this algorithm in simulations.The results show that when positioning error is 0%,its proportion is 82%;when positioning error is within 6 m,its proportion is 12%;when positioning error is 6-10 m,its proportion is 6%.Average positioning error is 0.75 m,and time of online average positioning is 0.56 sonly.It can meet the requirements of robust high-precision indoor traffic positioning.
In order to solve problems of indoor traffic positioning in GPS blind area,a high-precision method is proposed by combining Wi-Fi fingerprint localization with visual localization.In offline stage,Wireless AP MAC address and sample point coordinates are collected to establish a database of Wi-Fi location fingerprint,and gather plate images to develop training image sets,then calculate the SURF and ORB global feature descriptor of the training images,a visual positioning database is developed with the sample point coordinates.In positioning stage,undetermined measured fingerprint images and tested images are collected.A matching algorithm for fingerprints is used to get regional scopes and coordinates,and a matching algorithm for image features is used to obtain training or matching images with KNN algorithm,by querying orientation of visual database for visual range and coordinates.When the coordinates of the matching images are located within the range of Wi-Fi location,the visual location coordinates will be used as reference coordinates for final location,or vice versa.Different indoor traffic environments are applied to verify this algorithm in simulations.The results show that when positioning error is 0%,its proportion is 82%;when positioning error is within 6 m,its proportion is 12%;when positioning error is 6-10 m,its proportion is 6%.Average positioning error is 0.75 m,and time of online average positioning is 0.56 sonly.It can meet the requirements of robust high-precision indoor traffic positioning.
引文
[1]高伟,侯聪毅,许万旸,等.室内导航定位技术研究进展与展望[J].导航定位学报,2019,7(1):10-17.GAO Wei,HOU Congyi,XU Wanyang,et al.Research progress and prospect of indoor navigation and positioning technology[J].Journal of Navigation and Positioning,2019,7(10):17.(in Chinese)
[2]陶倩文,胡钊政,黄刚,等.基于消防安全疏散标志的高精度室内视觉定位[J].交通信息与安全,2018,36(2):39-46,60.TAO Qianwen,HU Zhaozheng,HU Gang,et al.High precision indoor visual positioning based on fire safety evacuation signs[J].Journal of Transport Information andSafety,2018,36(2):39-46,60.(in Chines)
[3]YASSIN A,NASSER Y,et al.Recent advances in indoor localization:A survey on theoretical approaches and applications[J].IEEE Communications Surveys and Tutorials,2017,19(2):1327-1346.
[4]曹凤萍,王荣本,张亮修.基于双目立体视觉的月球车三维定位算法[J].交通信息与安全,2012,30(4):28-33.CAO Fengpin,WANG Rongben,ZHANG Liangxiu,The three-dimensional localization algorithm of lunar rover based on binocular stereo vision[J].Journal of Transport Information and Safety,2012,30(4):28-33.(in Chinese)
[5]黄鹤,仇凯悦,李维,等.基于粒子滤波联合算法的室内地磁定位精度研究[J].西南交通大学学报,2019,5(2):1-7.HUANG He,CHOU Kairui,LI Wei,et al.Research on indoor geomagnetic positioning accuracy based on particle filter joint alorithm[J].Journal of Southwest Jiaotong University,2019,5(2):1-7.(in Chinese)
[6]GARCIAE,POUDEREUX P,et al.A robust UWB indoor Positionig system for highly complex environments[C].2015IEEE International Conference on Industril Technology International Conference on Industrial Technology,Seville,Spain:IEEE,2015.
[7]HULTGREN M,PAPATHANOPOULOS D.Evaluating the usage of bluetooth low energy beacons and smartphones for indoor positioning systems[D].G9teborg,Sweden:University of Gothenburg,2014.
[8]HE S,CHAN S H G.Wi-Fi fingerprint-based indoor positioning:Recent advances and comparisons[J].IEEE Communications Surveys&Tutorials,2016,18(1):466-490.
[9]沈雪微,冯宁,张玲玲,等.以WiFi和ZigBee联合定位的消防灭火救援系统[J].物联网技术,2015,5(1):32-35.SHEN Xuewei,FENG Ning,ZHANG Lingling,et al.Fire extinguishing and rescue system based on WIFI and ZigBee[J].Internet of Things Technologies,2015,5(1):32-35.(in Chinese)
[10]DORTZ N L,GAIN F,ZETTERBERG P.WiFi fingerprint indoor positioning system using probability distribution comparison[C].2012IEEE International Conference on Acoustics,Kyoto,Japan Speech:IEEE,2012.
[11]EISA S,PEIXOTO J,MENESES F,et al.Removing useless APs and fingerprints from WiFi indoor positioning radio maps[C].International Conference on Indoor Positioning&Indoor Navigation,Montbeliard France:IEEE,2014.
[12]石高涛,王伯远,吴斌.基于WiFi与移动智能终端的室内定位方法综述[J].计算机工程,2015,41(9):39-44.SHI Gaotao WANG Boyuan,WU Bin,An overview of indoor positioning methods based on WiFi and mobile intelligent terminals[J].Computer Engineering,2015,41(9):39-44.(in Chinese)
[13]吴雨,杨力,王梦茹,等.基于Android平台的WiFi定位系统研究与实现[J].全球定位系统,2016,41(4):90-93.WU Yu,YANG Li,WANG Mengru,et al.Research and implementation of WiFi location system based on Android platform[J].Gnss World of China,2016,41(4):90-93.(in Chinese)
[14]周牧,唐云霞,田增山,等.基于流形插值数据库构建的WLAN室内定位算法[J].电子与信息学报,2017,39(8):1826-1834.ZHOU Mu,TANG Yunxia,TIAN Zengshan,et al.WLAN indoor location algorithm based on manifold interpolation database[J].Journal of Electronics and Information Techn-o Logy,2017,39(8):1826-1834.(in Chinese)
[15]李祎承,胡钊政,胡月志,等.基于GPS与图像融合的智能车辆高精度定位算法[J].交通运输系统工程与信息,2017,17(3):112-119.LI Yicheng,HU Zhaozheng,HU Yuezhi,et al.High precision positioning algorithm of intelligent vehicle based on GPS and image fusion[J].Journal of Transportation Systems Engineering and Information Technology,2017,17(3):112-119.(in Chinese)
[16]DONG J,XIAO Y,NOREIKIS M,et al.iMoon:using smartphones for image-based indoor navigation[C].ACM Conference on Embedded Networked Sensor Systems,Seoul,South Korea:Association for Computing Machinery,2015.
[17]刘国忠,胡钊政.基于SURF和ORB全局特征的快速闭环检测[J].机器人,2017,39(1):36-45.LIU Guozhong,HU Zhaozheng.Fast closed-loop detection based on SURF and ORB global features[J].Robot,2017,39(1):36-45.(in Chinese)
[18]RUBBLEE E,RABAUD V,KONOLIGE K,et al.ORB:An efficient alternative to SIFT or SURF[C].2011IEEE International Conference on Computer Vision,Barcelona Spoin:IEEE,2011
[2]陶倩文,胡钊政,黄刚,等.基于消防安全疏散标志的高精度室内视觉定位[J].交通信息与安全,2018,36(2):39-46,60.TAO Qianwen,HU Zhaozheng,HU Gang,et al.High precision indoor visual positioning based on fire safety evacuation signs[J].Journal of Transport Information andSafety,2018,36(2):39-46,60.(in Chines)
[3]YASSIN A,NASSER Y,et al.Recent advances in indoor localization:A survey on theoretical approaches and applications[J].IEEE Communications Surveys and Tutorials,2017,19(2):1327-1346.
[4]曹凤萍,王荣本,张亮修.基于双目立体视觉的月球车三维定位算法[J].交通信息与安全,2012,30(4):28-33.CAO Fengpin,WANG Rongben,ZHANG Liangxiu,The three-dimensional localization algorithm of lunar rover based on binocular stereo vision[J].Journal of Transport Information and Safety,2012,30(4):28-33.(in Chinese)
[5]黄鹤,仇凯悦,李维,等.基于粒子滤波联合算法的室内地磁定位精度研究[J].西南交通大学学报,2019,5(2):1-7.HUANG He,CHOU Kairui,LI Wei,et al.Research on indoor geomagnetic positioning accuracy based on particle filter joint alorithm[J].Journal of Southwest Jiaotong University,2019,5(2):1-7.(in Chinese)
[6]GARCIAE,POUDEREUX P,et al.A robust UWB indoor Positionig system for highly complex environments[C].2015IEEE International Conference on Industril Technology International Conference on Industrial Technology,Seville,Spain:IEEE,2015.
[7]HULTGREN M,PAPATHANOPOULOS D.Evaluating the usage of bluetooth low energy beacons and smartphones for indoor positioning systems[D].G9teborg,Sweden:University of Gothenburg,2014.
[8]HE S,CHAN S H G.Wi-Fi fingerprint-based indoor positioning:Recent advances and comparisons[J].IEEE Communications Surveys&Tutorials,2016,18(1):466-490.
[9]沈雪微,冯宁,张玲玲,等.以WiFi和ZigBee联合定位的消防灭火救援系统[J].物联网技术,2015,5(1):32-35.SHEN Xuewei,FENG Ning,ZHANG Lingling,et al.Fire extinguishing and rescue system based on WIFI and ZigBee[J].Internet of Things Technologies,2015,5(1):32-35.(in Chinese)
[10]DORTZ N L,GAIN F,ZETTERBERG P.WiFi fingerprint indoor positioning system using probability distribution comparison[C].2012IEEE International Conference on Acoustics,Kyoto,Japan Speech:IEEE,2012.
[11]EISA S,PEIXOTO J,MENESES F,et al.Removing useless APs and fingerprints from WiFi indoor positioning radio maps[C].International Conference on Indoor Positioning&Indoor Navigation,Montbeliard France:IEEE,2014.
[12]石高涛,王伯远,吴斌.基于WiFi与移动智能终端的室内定位方法综述[J].计算机工程,2015,41(9):39-44.SHI Gaotao WANG Boyuan,WU Bin,An overview of indoor positioning methods based on WiFi and mobile intelligent terminals[J].Computer Engineering,2015,41(9):39-44.(in Chinese)
[13]吴雨,杨力,王梦茹,等.基于Android平台的WiFi定位系统研究与实现[J].全球定位系统,2016,41(4):90-93.WU Yu,YANG Li,WANG Mengru,et al.Research and implementation of WiFi location system based on Android platform[J].Gnss World of China,2016,41(4):90-93.(in Chinese)
[14]周牧,唐云霞,田增山,等.基于流形插值数据库构建的WLAN室内定位算法[J].电子与信息学报,2017,39(8):1826-1834.ZHOU Mu,TANG Yunxia,TIAN Zengshan,et al.WLAN indoor location algorithm based on manifold interpolation database[J].Journal of Electronics and Information Techn-o Logy,2017,39(8):1826-1834.(in Chinese)
[15]李祎承,胡钊政,胡月志,等.基于GPS与图像融合的智能车辆高精度定位算法[J].交通运输系统工程与信息,2017,17(3):112-119.LI Yicheng,HU Zhaozheng,HU Yuezhi,et al.High precision positioning algorithm of intelligent vehicle based on GPS and image fusion[J].Journal of Transportation Systems Engineering and Information Technology,2017,17(3):112-119.(in Chinese)
[16]DONG J,XIAO Y,NOREIKIS M,et al.iMoon:using smartphones for image-based indoor navigation[C].ACM Conference on Embedded Networked Sensor Systems,Seoul,South Korea:Association for Computing Machinery,2015.
[17]刘国忠,胡钊政.基于SURF和ORB全局特征的快速闭环检测[J].机器人,2017,39(1):36-45.LIU Guozhong,HU Zhaozheng.Fast closed-loop detection based on SURF and ORB global features[J].Robot,2017,39(1):36-45.(in Chinese)
[18]RUBBLEE E,RABAUD V,KONOLIGE K,et al.ORB:An efficient alternative to SIFT or SURF[C].2011IEEE International Conference on Computer Vision,Barcelona Spoin:IEEE,2011