【作者】 李颖；

【导师】 魏宇；

【作者基本信息】 哈尔滨工程大学 ， 控制科学与工程， 2018， 硕士

【摘要】 随着全球卫星导航定位技术的快速崛起,基于室外的导航系统蓬勃发展,在交通运输、国防安全等领域应用广泛。但在室内环境中,由于建筑物干扰等因素,室内很难获取到较为精确的卫星导航数据。随着人类活动在室内环境中增多,各种室内导航技术出现并迅速发展,人类生活也越来越依赖于室内导航技术,但不同的室内导航技术均存在很多不足之处。为提高导航精度,降低系统成本,便于移动行人导航,考虑室内环境特性,采用基于行人航位推算(PDR)、WIFI和地理标识融合的导航定位系统,满足行人在复杂的室内环境中的导航定位需求。为充分利用PDR、WIFI和地理标识的互补优势,本文提出了基于扩展卡尔曼滤波(EKF)的融合算法。对于WIFI定位系统,在实时定位阶段通过模糊推理加权位置指纹算法获取WIFI定位位置,利用行人前一步融合的位置数据与WIFI异常检测定义可信区域,降低计算成本,提高WIFI定位精度。地理标识标记室内环境特性较为明确的位置点,与WIFI系统在实时定位前期预测量阶段共同组建离线指纹数据库。PDR定位系统分析用户姿态模型,解算行人步态和步伐,利用陀螺仪和加速度传感器确定基于EKF航向模型的用户航向。为降低PDR累积误差,实现连续室内导航定位,以PDR系统数据为状态量,WIFI和地理标识为量测量,实现基于EKF融合的导航算法,并在Android平台搭建移动客户端,构建移动终端室内导航系统。经过对实际室内环境数据采集与移动终端导航系统测试,并对实验数据加以分析,可以得出:基于PDR、WIFI和地理标识融合的室内定位系统相较于PDR、WIFI融合系统和PDR、地理标识融合系统而言定位精度提升显著,其定位误差可基本保证在1m以内,满足日常的基于行人的室内导航需求。更多还原

【Abstract】 With the rapid development of the global satellite navigation and positioning technology,the outdoor navigation system is widely used in the fields of transportation,national defense security and so on.But the indoor environment is difficult to obtain more accurate satellite navigation data because of the building interference and other factors.With the increase of human activities in the indoor environment,various indoor navigation technology appears and developed rapidly,human life is increasingly dependent on indoor navigation technology,but there are many shortcomings in different indoor navigation technologies..In order to improve navigation accuracy,reduce system cost,facilitate mobile pedestrian navigation,considering indoor environmental characteristics,a navigation and positioning system based on Pedestrian Dead Reckoning(PDR),WIFI and landmarks fusion is applied to satisfy the navigation and location requirements of pedestrians in complex indoor environment.In order to make full use of the complementary advantages of PDR,WIFI and landmarks,a fusion algorithm based on extended Kalman filter(EKF)is proposed in this paper.For the WIFI positioning system,in the real-time positioning stage,the WIFI location is obtained by fuzzy reasoning weighted location fingerprint algorithm.The trusted area is defined by the location data and the WIFI anomaly detection of pedestrians before,so it can reduce the computation cost and improve the positioning accuracy of WIFI.The indoor environment location point is marked by landmarks which characteristics is more clear,and the off-line fingerprint database is formed together with the WIFI system in the phase of real-time positioning prediction.The PDR positioning system analyzes the user attitude model,calculates the pedestrian gait and steps,and uses the gyroscope and acceleration sensors to determine the user heading based on the EKF heading model.In order to reduce the accumulative error of PDR and achieve continuous indoor navigation and location,the navigation algorithm based on EKF fusion is implemented based on PDR system data as state variables,WIFI and landmarks as measure variables,and mobile client which is built on Android platform is built in the indoor navigation system for mobile terminals.After the test of the actual indoor environment data acquisition and mobile terminal navigation system,from the experimental data are analyzed,we can conclude that: the indoor positioning system PDR,WIFI and landmarks fusion compared to PDR and WIFI fusion system,PDR and landmarks system fusion system,the positioning accuracy is improved,and the precision of fusion was decided to based on the positioning error can be guaranteed about 1m,which is fitted the needs of pedestrian indoor navigation based on personal daily.更多还原