Performance enhancement for a GPS vector-tracking loop utilizing an adaptive iterated extended Kalman filter

Area of Science:

• * Navigation Systems
• * Signal Processing
• * State Estimation

Background:

• * Software-defined Global Positioning System (GPS) receivers face state estimation challenges in vector-tracking loops.
• * Nonlinear systems with model errors and white Gaussian noise require advanced filtering techniques.
• * Existing methods like Extended Kalman Filter (EKF), Adaptive Extended Kalman Filter (AEKF), and Iterated Extended Kalman Filter (IEKF) have limitations.

Purpose of the Study:

• * To propose a novel Adaptive Iterated Kalman Filter (AIEKF) for enhanced state estimation in GPS receivers.
• * To address model errors and white Gaussian noise in nonlinear systems.
• * To evaluate the performance of AIEKF in real-world road tests.

Main Methods:

• * Development of a noise statistics estimator to identify model errors.
• * Implementation of an Adaptive Iterated Kalman Filter (AIEKF) integrating the noise estimator.
• * Performance evaluation using a vector-tracking GPS receiver in road tests.

Main Results:

• * AIEKF demonstrates a significant accuracy advantage in position determination compared to IEKF and AEKF.
• * The proposed method effectively reduces the root-mean-square error (RMSE) for longitude, latitude, and altitude.
• * AIEKF achieved RMSE reductions of up to 54.6% (vs. EKF), 35.7% (vs. IEKF), and 30.7% (vs. AEKF) in different directions.

Conclusions:

• * The AIEKF is an effective approach for state estimation in vector-tracking GPS receivers.
• * The AIEKF significantly improves position accuracy and reduces RMSE.
• * The proposed method offers a superior alternative to existing Kalman filter variants for GPS applications.