融合RSSI和IMU数据的高可靠性定位方法
发布时间：2018-11-15 14:39
【摘要】：传感器技术及移动互联网的飞速发展促进位置服务逐渐渗透到了人类活动的各个方面,位置信息由此成为生活中至关重要的组成部分。几十年来,随着室外定位技术的日趋成熟,人们对于定位技术的探索逐渐转到室内。尽管出现了多种定位技术,但至今仍然缺乏一种高精度、高可靠的解决方案。目前使用规模最广的室内定位方法主要为指纹定位和PDR(Pedestrian Dead Reckoning)定位,但是无线信号容易受周围环境的影响,导致指纹定位的精度较低,性能不稳定;而PDR定位由于无法获知起始位置及误差累积的影响也难当大任。本文针对指纹定位精度低、性能差两大问题,研究融合RSSI(Received Signal Strength Indicator)及IMU(Inertial Measurement Unit)数据的高可靠性定位方法,获得了性能稳定的高精度定位体验。本文的研究成果主要如下:优化了指纹定位算法。通过研究设备兼容性及最佳扫描间隔的设定优化了数据采集,减轻了定位数据异常对指纹匹配的影响;然后根据RSSI的统计特性提出了加权距离模型,成功提高了指纹匹配的成功率;并利用信号源位置实现了定位结果的自动校正。实验证明优化后的指纹定位算法显著提高了定位精度。设计了一种粒子滤波融合定位算法。使用优化后的指纹定位输出生成粒子群,通过计算粒子位移与PDR位移的相似度求得粒子权值,从而限制定位结果的异常跳动,动态提升了定位算法的稳定性;并提出了一种误差反馈机制,避免了误差累积的影响。实际环境下的实验结果表明融合定位算法较好地反映了真实的运动轨迹。同时,定位精度及稳定性都得到了改善。
[Abstract]:With the rapid development of sensor technology and mobile Internet, location services have gradually penetrated into all aspects of human activities, so location information has become a vital part of life. In recent decades, with the maturation of outdoor positioning technology, the exploration of positioning technology has gradually shifted to indoor. Despite the emergence of a variety of positioning technology, but still lack of a high-precision, high-reliable solution. At present, the most widely used indoor positioning methods are fingerprint location and PDR (Pedestrian Dead Reckoning) location, but the wireless signal is easily affected by the surrounding environment, which leads to the low precision and unstable performance of fingerprint location. However, it is difficult for PDR location to know the starting position and the effect of error accumulation. Aiming at the problems of low precision and poor performance of fingerprint location, this paper studies a high reliability localization method combining RSSI (Received Signal Strength Indicator) and IMU (Inertial Measurement Unit) data, and obtains a high precision localization experience with stable performance. The main results of this paper are as follows: the fingerprint location algorithm is optimized. By studying the compatibility of the equipment and the setting of the best scanning interval, the data acquisition is optimized, and the influence of the abnormal location data on the fingerprint matching is reduced. Then according to the statistical characteristics of RSSI, a weighted distance model is proposed, which successfully improves the success rate of fingerprint matching, and realizes the automatic correction of location results by using the location of the signal source. The experimental results show that the optimized fingerprint location algorithm improves the location accuracy significantly. A particle filter fusion localization algorithm is designed. Using the optimized fingerprint location output to generate particle swarm, the particle weight value is obtained by calculating the similarity between particle displacement and PDR displacement, which limits the abnormal runout of the localization results and dynamically improves the stability of the localization algorithm. An error feedback mechanism is proposed to avoid the effect of error accumulation. The experimental results show that the fusion algorithm can well reflect the real motion trajectory. At the same time, the positioning accuracy and stability are improved.
【学位授予单位】：武汉大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN713
本文编号：2333586
[Abstract]:With the rapid development of sensor technology and mobile Internet, location services have gradually penetrated into all aspects of human activities, so location information has become a vital part of life. In recent decades, with the maturation of outdoor positioning technology, the exploration of positioning technology has gradually shifted to indoor. Despite the emergence of a variety of positioning technology, but still lack of a high-precision, high-reliable solution. At present, the most widely used indoor positioning methods are fingerprint location and PDR (Pedestrian Dead Reckoning) location, but the wireless signal is easily affected by the surrounding environment, which leads to the low precision and unstable performance of fingerprint location. However, it is difficult for PDR location to know the starting position and the effect of error accumulation. Aiming at the problems of low precision and poor performance of fingerprint location, this paper studies a high reliability localization method combining RSSI (Received Signal Strength Indicator) and IMU (Inertial Measurement Unit) data, and obtains a high precision localization experience with stable performance. The main results of this paper are as follows: the fingerprint location algorithm is optimized. By studying the compatibility of the equipment and the setting of the best scanning interval, the data acquisition is optimized, and the influence of the abnormal location data on the fingerprint matching is reduced. Then according to the statistical characteristics of RSSI, a weighted distance model is proposed, which successfully improves the success rate of fingerprint matching, and realizes the automatic correction of location results by using the location of the signal source. The experimental results show that the optimized fingerprint location algorithm improves the location accuracy significantly. A particle filter fusion localization algorithm is designed. Using the optimized fingerprint location output to generate particle swarm, the particle weight value is obtained by calculating the similarity between particle displacement and PDR displacement, which limits the abnormal runout of the localization results and dynamically improves the stability of the localization algorithm. An error feedback mechanism is proposed to avoid the effect of error accumulation. The experimental results show that the fusion algorithm can well reflect the real motion trajectory. At the same time, the positioning accuracy and stability are improved.
【学位授予单位】：武汉大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN713
【参考文献】
相关期刊论文 前3条
1 郑学理;付敬奇;;基于PDR和RSSI的室内定位算法研究[J];仪器仪表学报;2015年05期
2 程士安;陈思;;基于地理位置服务(LBS)技术平台的传播规律——以“街旁”为例解读技术赋予信息分享的新权力[J];新闻大学;2010年04期
3 方震;赵湛;郭鹏;张玉国;;基于RSSI测距分析[J];传感技术学报;2007年11期
，本文编号：2333586
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/2333586.html
