Reactive Autonomous Navigation of UAVs for Dynamic Sensing Coverage of Mobile Ground Targets

Area of Science:

• Robotics
• Artificial Intelligence
• Computer Vision

Background:

• Autonomous navigation of unmanned aerial vehicles (UAVs) is crucial for surveillance tasks.
• Tracking multiple moving ground targets presents challenges due to dynamic environments and terrain variations.

Purpose of the Study:

• To develop and evaluate an autonomous navigation algorithm for UAVs to effectively surveil multiple moving ground targets.
• To enhance sensing coverage and minimize target revisit times using a reactive control strategy and Voronoi partitioning.

Main Methods:

• A reactive real-time sliding mode control algorithm guides communicating UAVs equipped with video cameras.
• The Voronoi partitioning technique is integrated to optimize UAV movement and reduce target revisit intervals.
• Extensive computer simulations are performed, including scenarios with varying numbers of UAVs, targets, and uneven terrain.

Main Results:

• The proposed algorithm successfully navigates UAVs towards moving targets to maximize sensing coverage.
• Voronoi partitioning significantly reduces target revisit times compared to methods without this technique.
• Simulations demonstrate the algorithm's effectiveness across diverse scenarios, including complex multi-UAV and multi-target situations on uneven ground.

Conclusions:

• The developed autonomous navigation algorithm provides an effective solution for multi-target surveillance using UAVs.
• Integrating Voronoi partitioning enhances surveillance efficiency by reducing revisit times, despite a slight increase in computational load.
• The approach is robust and adaptable to different terrain conditions, making it suitable for real-world applications.