Maximum correentropy-based robust Square-root Cubature Kalman Filter for vehicular cooperative navigation

Area of Science:

• Robotics and Autonomous Systems
• Signal Processing
• Navigation and Control

Background:

• Relative positioning is crucial for intelligent vehicle navigation and network collaboration.
• High measurement noise contamination severely degrades traditional filtering performance.

Purpose of the Study:

• To develop a robust vehicular cooperative localization method resilient to noise contamination.
• To improve the accuracy and reliability of relative position estimation in intelligent vehicles.

Main Methods:

• Proposed a novel Maximum Correentropy Robust Square-root Cubature Kalman Filter (MCSCKF).
• Leveraged the robustness of Maximum Correentropy while retaining Square-root Cubature Kalman Filter (SCKF) advantages.
• Evaluated the method in tightly integrated vehicular cooperative navigation scenarios.

Main Results:

• MCSCKF demonstrated superior performance compared to Extended Kalman Filter (EKF) and Cubature Kalman Filter (CKF).
• Localization accuracy improved by 35.08% over EKF and 31.83% over CKF.
• The proposed filter exhibits strong robustness against non-Gaussian noise.

Conclusions:

• The MCSCKF is an effective method for enhancing vehicular cooperative localization accuracy and robustness.
• This approach addresses the performance degradation issues caused by measurement noise contamination.
• MCSCKF offers a promising solution for reliable intelligent vehicle navigation.