Multi-UAV Path Planning for Autonomous Missions in Mixed GNSS Coverage Scenarios

Area of Science:

• Robotics
• Aerospace Engineering
• Computer Science

Background:

• Navigation in challenging environments like urban canyons requires robust strategies for unmanned aerial vehicles (UAVs).
• Heterogeneous Global Navigation Satellite Systems (GNSS) coverage presents unique challenges for multi-UAV coordination and path planning.
• Existing methods may not efficiently adapt to dynamic changes in navigation conditions.

Purpose of the Study:

• To develop an adaptive algorithm for multi-UAV path planning in environments with mixed GNSS availability.
• To enable autonomous reconfiguration of multi-UAV systems based on real-time mission requirements and environmental conditions.
• To optimize flight trajectories for efficient coexistence and reduced mission time.

Main Methods:

• Formulating path planning as a vehicle routing problem to generate smooth, polynomial trajectories.
• Estimating flight times to ensure safe UAV coexistence in GNSS-degraded areas.
• Implementing a distributed system that autonomously adapts to varying navigation conditions.
• Testing the algorithm in a realistic 3D simulation environment with variable numbers of UAVs and waypoints.

Main Results:

• The algorithm demonstrates computational efficiency, with burden largely independent of the number of UAVs.
• Near real-time implementation is feasible, even with a significant number of waypoints.
• The solution effectively utilizes available flight resources, reducing overall mission duration.
• The system successfully manages cooperative and independent flight modes based on GNSS availability.

Conclusions:

• The proposed algorithm offers an efficient and adaptive solution for multi-UAV path planning in heterogeneous GNSS environments.
• It provides a robust framework for autonomous system reconfiguration, enhancing operational flexibility.
• The method optimizes resource utilization and mission time, making it suitable for practical applications.