Heading constraint algorithm for foot-mounted PNS using low-cost IMU

Foot-mounted pedestrian navigation system (PNS) is a common solution to using micro-electro mechanical (MEMS) inertial sensors. The inherent problems of (INS) by the traditional algorithm, such as accumulated errors and lack observation heading altitude information, have become obstacles application development PNS. In this paper, we introduce heuristic constraint method. First all, according movement characteristics human gait, use generalized likelihood ratio test (GLRT) detector time threshold classify so that can effectively identify stationary state foot. addition, based on zero velocity update (ZUPT) angular rate (ZARU), cumulative error measurement unit (IMU) limited corrected, then estimation used constrain correct pedestrian. After simulation experiments with low-cost IMU, method proved reduce localization end-point less than 1% total distance, it has great value in application.