Tightly Coupled GNSS/INS Integration with Robust Sequential Kalman Filter for Accurate Vehicular Navigation

Area of Science:

• Navigation Systems Engineering
• Geomatics Engineering
• Robotics and Autonomous Systems

Background:

• Advancements in multi-constellation, multi-frequency Global Navigation Satellite Systems (GNSS) provide increased observational data for GNSS/Inertial Navigation System (INS) integration.
• Existing integration methods face challenges in accuracy, robustness, and computational load.

Purpose of the Study:

• To develop a robust and computationally efficient tightly coupled GNSS/INS integration method.
• To maximize navigation accuracy using pseudorange, Doppler, and carrier phase observations simultaneously.
• To enhance fault detection and adaptation in GNSS channels.

Main Methods:

• A novel tight integration model utilizing pseudorange, Doppler, and carrier phase observations.
• Sequential Kalman Filter (KF) for efficient processing of high-dimensional observation vectors.
• A robust estimation method employing Gaussian testing for GNSS channel fault detection.

Main Results:

• Significant improvements in velocity and attitude accuracy compared to loose and conventional tight coupling methods (up to 69.42% and 47.16%).
• Enhanced computational efficiency by approximately 53.09% compared to batch KF processing.
• Superior performance in challenging GNSS environments, including better bridging capability during outages.

Conclusions:

• The proposed method offers a robust and accurate solution for real-time vehicular navigation.
• It effectively addresses accuracy, robustness, and computational efficiency challenges in GNSS/INS integration.
• Demonstrates significant performance gains in both favorable and challenging GNSS conditions.