Polarized light compass-aided inertial navigation under discontinuous observations environment

Area of Science:

• Navigation systems
• Sensor fusion
• Robotics

Background:

• Polarized light/inertial heading measurement systems are crucial for autonomous platforms when satellite navigation is unavailable.
• Existing systems suffer from noise in complex environments (e.g., urban areas) and signal loss in underpasses, impacting accuracy and reliability.
• Heading accuracy is critical for unmanned combat platform navigation and operation.

Purpose of the Study:

• To develop robust algorithms for integrated polarized light/inertial heading measurement systems.
• To address limitations of existing systems, specifically noise interference and signal loss in challenging environments.
• To enhance the precision and stability of autonomous heading measurements for unmanned platforms.

Main Methods:

• A robust Cubature Kalman filter (CKF) data-fusion algorithm was proposed to mitigate sharp noise from environmental obstructions.
• A random forest regression (RFR) neural network model was developed to handle discontinuous polarized light compass signals, such as in underpasses.
• The algorithms were verified through simulated and outdoor experiments using vehicle-mounted data.

Main Results:

• The robust CKF algorithm reduced the Root Mean Square Error (RMSE) of heading angle to 0.3612° in complex environments.
• The RFR model achieved an RMSE of 1.1894° for heading angle, successfully utilizing discontinuous observations.
• The combined system demonstrated improved precision and stability compared to traditional polarized light/inertial systems.

Conclusions:

• The proposed robust CKF algorithm effectively filters poor measurements, enhancing heading accuracy in environments with trees and tall buildings.
• The RFR model successfully addresses the issue of complete polarized light compass signal loss in underpasses, enabling continuous heading detection.
• These advancements significantly improve the performance and reliability of polarized light/inertial heading measurement systems for autonomous platforms in challenging conditions.