ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술지 An Algorithm on Gyro Drift Elimination in Personal Navigation System for Firefighter Localization with a Waist-Worn MEMS IMU Sensor System
Cited 2 time in scopus Download 12 time Share share facebook twitter linkedin kakaostory
문지혜, 양회성, 이강복, 명승일
제어로봇시스템학회논문지, v.25 no.1, pp.88-97
18ZH1100, 사물-사람-공간의 유기적 연결을 위한 초연결 공간의 분산 지능 핵심원천 기술, 손영성
In this paper, we propose five techniques applying the Teager-Kaiser energy operator (TKEO) to eliminate gyro drifts with a waist-worn IMU sensor for infra-less pedestrian dead reckoning (PDR). For providing the service to firefighters in a variety of environments, in this research, we selected a low-cost IMU sensor and used acceleration and gyro. The PDR consists of walking distance and direction estimation processes; in the distance estimation process, the peak and zero crossing detection algorithms were used to detect the steps, and the walking distance was calculated using a fuzzy interface system. For estimating the correct direction, the roll and pitch drifts were eliminated by using an extended Kalman filter. In the process, we developed a yaw drift reduction algorithm that consists of two filters and devised three techniques with TKEO. The drift was modeled as a linear function and reduced with the three techniques, and the extra noises were eliminated by the filters. To compare the existing work, the heuristic drift reduction (HDR) was implemented. In the experiments, after each walk along 580 m and 234 m, the proposed algorithm reported position errors of 0.37% and 0.29%, which exceeded the 2.31% and 6.14% of the HDR.
Extended Kalman Filters, Gyro drift, Pedestrian dead reckoning, Teager-Kaiser energy operator, Waist-worn IMU
KSP 제안 키워드
Detection algorithm, Direction estimation, Distance and direction, Drift elimination, Drift reduction, Extended kalman fiLTEr, Fuzzy interface system(FIS), Gyro drift, IMU sensor, Linear function, Low-cost