基于人体多方位运动的三维自主导航定位算法  

Three-dimensional autonomous navigation algorithm based on multi-azimuth movement pattern

作　　者：刘宇[1] 周帆[1] 李云梅[1] 路永乐[1] 彭慧[2] 张欣[1] LIU Yu ZHOU Fan LI Yun-mei LU Yong-le PENG Hui ZHANG Xin(Chongqing Municipal Level Key laboratory of Photoelectronic Information Sensing and Transmitting Technology, Chongqing University of Post and Telecommunications, Chongqing 400065, China The 26th Institute of China Electronics Technology Corporation, Chongqing 400065, China)

机构地区：[1]重庆邮电大学光电信息感测与传输技术重庆市重点实验室,重庆400065 [2]中国电子科技集团第二十六研究所,重庆400065

出　　处：《中国惯性技术学报》2016年第4期449-453,共5页Journal of Chinese Inertial Technology

基　　金：国家自然科学基金资助项目(51175535);国际联合研究中心科技平台与基地建设(cstc2014gjhz0038);重庆市基础与前沿研究计划项目(cstc2015jcyj BX0068)

摘　　要：基于MEMS传感器的行人导航通常利用行人航迹推算算法解算出人体位置的坐标对行人定位。传统行人航迹推算算法只能用于单一前进行走运动模式,不适用于人体实际的多方位运动模式。因此,提出了一种多方位运动三维自主导航定位算法,利用三轴加速度计数据对行人进行步态检测和步长估计,在分析了加速度计单轴数据的基础上,讨论了4类运动模式的识别方法,推导出适用于人体多方位运动模式的航迹推算公式。在智能手机平台上进行测试验证,实验结果表明,所提出的多方位运动模式自主导航定位算法在人体实际运动中,定位误差小于3.2%,很大程度上提高了导航精度。The pedestrian navigation based on MEMS sensor usually achieves real-time pedestrian positioning by using pedestrian dead reckoning（PDR） algorithm to calculate the coordinate of pedestrian＇s position. Traditional PDR only considers forward movement pattern, which is not applicable to the actual multi-azimuth human movement pattern. Hence, a novel algorithm of three-dimensional autonomous navigation based on multi-azimuth movement pattern is proposed by using the data of three-axis accelerometer to implement step detection and stride-length estimation. Based on the analysis of the accelerometer data, a pattern recognition method for four types of movement patterns is discussed, and the dead reckoning equation applied to pedestrian multi-azimuth movement pattern is derived. Verification experiments are made on smartphone platform by using the proposed algorithm into the actual autonomous navigation based on multi-azimuth movement pattern, which show that the positioning error is less than 3.2%, significantly improving the navigation accuracy.

关 键 词：MEMS传感器 三维自主导航 航迹推算 多方位 

分 类 号：U666.1[交通运输工程—船舶及航道工程]