Combined Location Method Based on Error State Kalman Filter

Abstract

Based on the data fusion method of iterative Error State Kalman Filter framework, this paper realizes the combined location of optical and inertial navigation. The method uses a tight coupling combination, uses the measurement value of Optical Position Sensor (OPS) to the marked point to correct the prior estimation of IMU, and takes the data error of OPS and IMU as the observation and state variables, thus simplifying the calculation of Jacobian matrix and avoiding the complexity caused by nonlinearity. The experiment shows that this method can effectively improve the positioning accuracy, enhance the robustness of OPS, and has good practicability.