Robust Target Tracking with Multi-Static Sensors under Insufficient TDOA Information.

Area of Science:

• Acoustics and Signal Processing
• Underwater Navigation and Tracking

Background:

• Underwater target tracking relies on multi-static sonar networks with passive sonobuoys and active pings.
• Estimating target location uses Time Difference of Arrival (TDOA) measurements.
• Harsh underwater conditions like ambient noise lead to insufficient and inaccurate TDOA data, degrading tracking performance.

Purpose of the Study:

• To develop a robust target tracking algorithm for multi-static sonar networks operating in challenging underwater environments.
• To mitigate performance degradation caused by limited and erroneous TDOA measurements.

Main Methods:

• Implemented track management with track splitting to handle insufficient measurements.
• Utilized a Gaussian Mixture Model (GMM) for fusing multiple TDOA measurements to estimate target location.
• Employed a stack-based data association method using multi-frame measurements to refine target trajectory.

Main Results:

• The proposed algorithm demonstrates effectiveness in improving underwater target tracking accuracy.
• Track splitting successfully reduces performance degradation from sparse measurements.
• GMM-based fusion and stack-based data association enhance trajectory estimation precision.

Conclusions:

• The developed robust target tracking algorithm effectively addresses challenges posed by noisy underwater environments.
• The combination of track management, GMM fusion, and advanced data association provides a significant improvement in tracking performance.
• Simulation results validate the proposed method's capability for accurate underwater target localization and trajectory estimation.