Adaptive Kalman Filter-Based Single-Beacon Underwater Tracking with Unknown Effective Sound Velocity

Area of Science:

• Underwater acoustics
• Navigation systems
• Signal processing

Background:

• Single-beacon systems use acoustic range measurements to correct dead reckoning errors.
• Accurate estimation of effective sound velocity (ESV) is crucial but challenging, leading to positioning inaccuracies.
• Extended Kalman Filters (EKF) struggle with unknown and time-varying noise characteristics in underwater environments.

Purpose of the Study:

• To develop an improved single-beacon underwater tracking algorithm.
• To address the limitations of EKF in handling uncertain acoustic propagation and ESV.
• To enhance the accuracy and robustness of underwater vehicle positioning.

Main Methods:

• Proposed a novel "5-sv" model treating unknown ESV as a state variable.
• Developed an adaptive Kalman filter (AKF) algorithm based on the "5-sv" model.
• Validated the AKF using simulated and field data, comparing filter and smoother results.

Main Results:

• The proposed AKF algorithm significantly improves tracking accuracy compared to traditional methods.
• The AKF effectively estimates unknown noise parameters, which align well with true values.
• The adaptive approach mitigates divergence issues often encountered with EKF.

Conclusions:

• The adaptive Kalman filter offers a robust solution for single-beacon underwater tracking.
• Accurate estimation of ESV and noise parameters is vital for precise underwater navigation.
• The developed algorithm provides a significant advancement in underwater positioning technology.