동적 환경에서 칼만필터 기반 GNSS 위치 오류 검출 기법 연구

초록

GNSS(Global Navigation Satellite System) is widely used in autonomous vehicles, drones, and MUM-T(Manned-Unmanned Teaming) systems. However, positioning errors frequently occur in real environments due to satellite geometry changes and signal interruptions. This study proposes a real-time GNSS error detection and correction framework by integrating a multi-GNSS fusion algorithm with a Kalman filter-based error detection algorithm. The proposed method estimates corrected positions using the Weighted Least Squares(WLS) method based on position and error covariance of individual satellite constellations. Position reliability is evaluated using the Normalized Innovation Squared(NIS) derived from the Kalman filter, and a time synchronization structure for multiple GNSS receivers is applied. Driving experiments using an autonomous vehicle platform demonstrated that the proposed algorithm could detect abnormal positioning situations in real time and improve positioning stability in environments with intermittent signal interruptions and outliers. The proposed framework is expected to be applicable to autonomous vehicles and MUM-T systems requiring reliable positioning performance.

키워드