Tracking a Decentralized Linear Trajectory in an Intermittent Observation Environment.

Area of Science:

• Control Systems Engineering
• Signal Processing
• Robotics and Automation

Background:

• Decentralized tracking systems are susceptible to sensor faults and noise, complicating data processing.
• Existing fault detection schemes primarily address faults, not the combined impact of faults and noise.
• Noise significantly degrades sensor communication and processing, exacerbating tracking inaccuracies.

Purpose of the Study:

• To develop and evaluate a robust tracking scheme for decentralized systems operating under both sensor faults and noise.
• To investigate the performance of a Kalman filter-based approach in mitigating the adverse effects of simultaneous faults and noise.
• To address the challenges posed by local single and multiple faults in the presence of system noise.

Main Methods:

• Integration of a Kalman filter with existing fault detection and isolation schemes.
• Development of a general scenario for decentralized tracking systems.
• Case study involving a target tracking scenario, with and without noise.
• Testing the proposed schemes against various types of faults.

Main Results:

• The proposed Kalman filter-based scheme demonstrates effective tracking performance in decentralized systems with simultaneous faults and noise.
• The approach successfully mitigates the negative impacts of noise on sensor data processing and communication.
• Experimental results confirm the acceptable performance and reliability of the developed tracking scheme.

Conclusions:

• The Kalman filter provides a viable solution for enhancing the robustness of decentralized tracking systems against combined faults and noise.
• The presented methodology offers a significant improvement over existing methods that do not account for noise.
• The developed scheme shows promise for practical applications requiring reliable tracking in noisy and fault-prone environments.