Modified Artificial Potential Field for the Path Planning of Aircraft Swarms in Three-Dimensional Environments

Area of Science:

• Robotics
• Artificial Intelligence
• Control Systems

Background:

• Path planning is critical for autonomous systems, requiring consideration of safety and computational efficiency.
• Existing methods often struggle with local minima and oscillations in obstacle avoidance.
• Real-time path planning in dynamic environments with multiple agents remains a challenge.

Purpose of the Study:

• To introduce a modified potential field method for robust path planning in autonomous systems.
• To address and overcome limitations of traditional potential field methods, specifically local minima and oscillations.
• To enable autonomous systems, such as robotic swarms, to navigate complex dynamic environments safely and efficiently.

Main Methods:

• A modified potential field method incorporating a three-dimensional (3D) vortex field.
• Independent and automatic direction selection for the vortex field based on robot-object proximity.
• Implementation and testing on Crazyflie 2.0 aircraft with a loco positioning system for state estimation.

Main Results:

• The proposed method effectively provides obstacle avoidance capabilities.
• Local minima problems and oscillations within repulsive fields were successfully eliminated.
• Feasible paths were generated in real-time for swarm flight and moving target pursuit scenarios.

Conclusions:

• The modified potential field method offers a robust solution for real-time path planning in autonomous systems.
• The 3D vortex field strategy enhances navigation safety and efficiency in dynamic, multi-agent environments.
• The approach is experimentally validated, demonstrating its practical applicability for complex robotic tasks.