A novel multi-objective dung beetle optimizer for Multi-UAV cooperative path planning.

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Operations Research

Background:

• Multi-unmanned aerial vehicle (UAV) path planning is vital for collaborative missions.
• Traditional methods struggle with convergence, diversity, and constraint handling.
• Complex multi-objective optimization is inherent in collaborative UAV path planning.

Purpose of the Study:

• To develop an advanced algorithm for multi-UAV path planning.
• To address limitations of existing methods in convergence, diversity, and constraint management.
• To optimize collaborative path planning considering multiple objectives.

Main Methods:

• Introduced a directional evolutionary non-dominated sorting dung beetle optimizer with adaptive stochastic ranking (DENSDBO-ASR).
• Formulated two objectives: total cost (length, altitude) and total cost (threat, time).
• Integrated directional mutation and crossover for enhanced convergence and global search, plus adaptive stochastic ranking for constraint handling.

Main Results:

• DENSDBO-ASR demonstrated superior performance in convergence accuracy and population diversity compared to five other algorithms.
• Successfully identified feasible paths in challenging scenarios, achieving low objective function values (e.g., 637.26 and 0).
• Effectiveness validated through constrained problem functions (CF) tests, Wilcoxon rank sum test, and Friedman test.

Conclusions:

• DENSDBO-ASR offers an exceptional optimization approach for multi-UAV path planning challenges.
• The algorithm effectively balances convergence and diversity while managing complex constraints.
• Proven superiority makes DENSDBO-ASR a valuable tool for collaborative robotic operations.