Variable-Speed UAV Path Optimization Based on the CRLB Criterion for Passive Target Localization.

Area of Science:

• Robotics and Control Systems
• Signal Processing
• Optimization Algorithms

Background:

• Passive target localization accuracy heavily depends on unmanned aerial vehicle (UAV) swarm configuration.
• Existing methods often assume fixed UAV speeds, limiting feasible positioning strategies.

Purpose of the Study:

• To investigate UAV path optimization for improved passive target localization accuracy.
• To address the challenge of variable UAV speeds in path planning.

Main Methods:

• Developed a passive target localization model using time difference of arrival (TDOA), enhanced by Chan method and Taylor series expansion.
• Derived Cramer-Rao lower bound (CRLB) for the modified TDOA algorithm to guide UAV positioning.
• Incorporated variable UAV speeds, expanding the feasible region to an annular shape.
• Applied particle swarm optimization (PSO) for efficient path optimization.

Main Results:

• The proposed method enhances localization accuracy by optimizing UAV positions within an annular feasible region.
• Variable speed consideration extends the operational range and improves localization performance.
• PSO algorithm effectively addresses the increased computational complexity of path optimization.

Conclusions:

• UAV path optimization significantly boosts passive target localization accuracy.
• The novel approach considering variable speeds and annular regions offers superior performance.
• PSO provides an efficient solution for real-time UAV path planning in localization tasks.