摘要
多种运动状态情况下,由于对步行者航位推算系统中步态检测不准确,导致推算定位误差增大。针对上述弊端,提出了多运动状态下自适应阈值步态检测算法。首先,通过改进的多阈值的步态检测算法,改善支撑区间检测的准确性。然后利用随机森林算法识别站立、走路、跑步、上楼和下楼等五种常见的室内运动状态,自适应匹配相应的步态检测的阈值,以实现多运动状态下的精准步态检测,提高行人航位推算系统精度。实测结果证明,相对于传统固定阈值检测算法,该算法平面定位精度提升33.1%,步态检测精度提升89.4%。
Under various motion states the gait detection in the pedestrian dead reckoning system is not always accurate,which leads to the increase of the positioning error.To solve the problem,an adaptive threshold gait detection algorithm in multi-motion state is proposed.Firstly,the improved multi-threshold gait detection algorithm is used to improve the accuracy of support interval detection.Secondly,the random forest algorithm is used to identify five common motions such as standing,walking,running,going upstairs and going downstairs,then the thresholds of corresponding gait detection is adaptively matched to the motions,so accurate gait detection in multiple motion states will be achieved and the precision of pedestrian dead reckoning(PDR)system will be improved.The experiment result shows that the horizontal positioning accuracy of the system used the proposed algorithm is improved by 33.1%,and the gait detection accuracy is improved by 89.4%,compared with the traditional algorithm.
作者
宁一鹏
梁建
王坚
NING Yipeng;LIANG Jian;WANG Jian(School of Surveying and Geo-Informatics,Shandong Jianzhu University,Jinan 250101,China;School of Environment Science and Spatial Information,China University of Mining and Technology,Xuzhou 221116,China;School of Geomatics and Urban Spatial Informatics,Beijing University of Civil Engineering and Architecture,Beijing 102616,China)
出处
《中国惯性技术学报》
EI
CSCD
北大核心
2020年第2期172-178,185,共8页
Journal of Chinese Inertial Technology
基金
国家自然科学基金青年基金(41904029)
2019年山东省高等学校青创人才引育计划。
关键词
行人航位推算
多阈值
自适应
步态检测
随机森林算法
pedestrian dead reckoning
multi-threshold
adaptive
gait detection
random forest algorithm
作者简介
宁一鹏(1990—),男,讲师,博士研究生,从事无缝定位技术研究。E-mail:ningyipeng19@sdjzu.edu.cn。
