A State Constraint Kalman Filter for Pedestrian Navigation with Low Cost MEMS Inertial Sensors

INS/GPS and INS/ZUPT techniques are complementary for pedestrian navigation applications. For optimal use of INS/ZUPT, the MEMS sensors are best mounted on a foot of a user, since the foot can periodically become stationary, thus applying ZUPT. For GPS/INS based module is best mounted on the upper body of the user, so that the user’s body does not block the GPS signals. In this paper, we propose a new solution for a pedestrian navigation scheme to fuse the information from the two systems. When both systems are used at the same time, a constraint that a spatial maximal distance of 1.5 m between the positions estimates of the two systems exists. Through exploring this external information, the state constraint Kalman filter was used to minimize the positioning errors of both systems. The experimental results show that when GPS is available, the INS/ZUPT sub-system’s positioning errors could be well bounded; on the other hand, when GPS signal is not available (simulated), the INS/GPS sub-system’s positioning accuracy could also be significantly improved.