Acoustic Target Tracking Through a Cluster of Mobile Agents

Area of Science:

• Robotics and Autonomous Systems
• Acoustic Signal Processing
• Sensor Networks

Background:

• Conventional fixed-array acoustic localization suffers from performance limitations.
• Tracking moving targets with mobile sensor networks presents challenges in coordination and adaptation.

Purpose of the Study:

• To develop an improved method for localizing and tracking moving acoustic targets using a cluster of mobile agents.
• To enhance target localization and tracking performance compared to fixed-array systems.

Main Methods:

• Modeling a mobile agent cluster as a sensor network for acoustic emission sensing.
• Proposing a centralized control strategy for dynamic cluster reconfiguration based on target position estimation.
• Utilizing a compatible set representation for future target positions to account for processing delays.
• Implementing a frame-by-frame cluster reconfiguration algorithm to maximize overlap between sensor sensitivity and target's compatible set.

Main Results:

• Simulations demonstrate effective localization and tracking of moving targets using acoustic emissions.
• The proposed approach shows significant improvements in localization and tracking performance compared to conventional static-array localization.
• Analysis includes the impact of the Doppler effect on performance for moving sources and sensors.

Conclusions:

• The mobile agent cluster approach offers superior acoustic target tracking capabilities.
• The dynamic reconfiguration strategy effectively adapts to target movement and optimizes sensor network sensitivity.
• The method is adaptable to other sensing modalities like video or multimodal information.