A derivative UKF for tightly coupled INS/GPS integrated navigation.

Area of Science:

• Navigation Systems Engineering
• Estimation Theory
• Signal Processing

Background:

• Inertial Navigation System/Global Positioning System (INS/GPS) integration is crucial for accurate navigation.
• Nonlinearity in Kalman filter measurement equations, arising from raw GPS pseudorange, poses challenges.
• Extended Kalman Filter (EKF) linearization can lead to modeling errors and degraded solutions.

Purpose of the Study:

• To develop a more accurate and computationally efficient method for INS/GPS integration.
• To address the limitations of EKF linearization and traditional UKF computational burden.
• To improve navigation estimation accuracy in INS/GPS systems.

Main Methods:

• Constructed a nonlinear measurement equation for INS/GPS integration by including second-order Taylor series terms.
• Developed a derivative Unscented Kalman Filter (UKF) that combines the Kalman Filter (KF) prediction with UKF update.
• Applied the derivative UKF to INS/GPS navigation estimation.

Main Results:

• The derivative UKF achieves higher navigation accuracy compared to the traditional UKF.
• The proposed derivative UKF significantly reduces computational cost, especially for high-dimensional systems.
• Theoretical analysis and simulations validate the effectiveness of the derivative UKF.

Conclusions:

• The derivative UKF offers a superior approach for INS/GPS integration, balancing accuracy and computational efficiency.
• This method effectively mitigates the drawbacks of EKF linearization and traditional UKF computational load.
• The derivative UKF presents a promising advancement for real-time navigation applications.