Multi-Strategy optimisation algorithm for multi-UAV path planning in complex mountainous environments.

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Optimization Algorithms

Background:

• Unmanned Aerial Vehicle (UAV) swarms require advanced path planning for complex missions.
• Existing algorithms often struggle with dynamic obstacle avoidance and global path optimization.

Purpose of the Study:

• To develop a novel three-dimensional path planning method for UAV swarms.
• To enhance obstacle avoidance capabilities and optimize flight paths using a new metaheuristic algorithm.

Main Methods:

• Proposed the Multi-Strategy Improved Crested Porcupine Optimizer (ICPO) for UAV swarm path planning.
• Integrated tent mapping and reverse refraction learning to improve population diversity.
• Incorporated animal escape behavior, moving escape strategy, and Cauchy distribution for enhanced global search and flexibility.
• Introduced Cauchy-Gaussian mutation to prevent local optima stagnation.

Main Results:

• ICPO demonstrated superior robustness and stability in simulations on 2005 and 2022 test sets.
• The algorithm proved effective for three-dimensional path planning in complex terrain simulations.
• Enhanced population diversity and global search capabilities were observed.

Conclusions:

• The proposed ICPO algorithm offers an effective solution for 3D path planning in UAV swarms.
• ICPO enhances UAV navigation by improving obstacle avoidance and optimizing flight trajectories.
• The method shows significant potential for real-world applications requiring autonomous aerial navigation.