Multi-UAV Cooperative Path Planning Using a Behavior-Adaptive Aquila Optimizer Under Multiple Constraints.

Area of Science:

• Artificial Intelligence
• Robotics
• Optimization Algorithms

Background:

• Multi-unmanned aerial vehicle (UAV) cooperative path planning faces challenges like high dimensionality, nonlinearity, and multiple constraints.
• Existing optimization algorithms may struggle with dynamic adjustments and stability in complex scenarios.

Purpose of the Study:

• To propose an enhanced Aquila Optimizer (EAO) for improved multi-UAV cooperative path planning.
• To enhance search performance, stability, and adaptability in dynamic environments.

Main Methods:

• Developed a Behavior-Adaptive Aquila Optimizer (EAO) with a multi-strategy cooperative framework.
• Integrated mechanisms for diversity maintenance, dynamic neighborhood guidance, differential evolution-based exploitation, and adaptive behavior selection.
• Validated EAO on CEC2017 and CEC2020 benchmark suites and applied it to multi-UAV path-planning simulations.

Main Results:

• EAO achieved the best overall performance ranking against 13 other algorithms on benchmark suites.
• In multi-UAV simulations, EAO demonstrated superior performance in path-planning fitness, effective trajectories, and runtime.
• EAO improved average fitness by up to 84.84% compared to the original Aquila Optimizer (AO).

Conclusions:

• The proposed Behavior-Adaptive Aquila Optimizer (EAO) is effective and feasible for multi-UAV cooperative path planning.
• EAO's dynamic behavior adjustment enhances search performance and stability in complex, constrained environments.
• EAO offers significant improvements over the standard Aquila Optimizer for UAV path planning.