Multi-strategy improved gazelle optimization algorithm for numerical optimization and UAV path planning.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• The Gazelle Optimization Algorithm (GOA) is a recognized metaheuristic but faces challenges with slow convergence and local optima.
• Practical applications require optimization algorithms with improved precision and faster convergence rates.

Purpose of the Study:

• To introduce a Multi-Strategy Improved Gazelle Optimization Algorithm (MIGOA) to overcome the limitations of the standard GOA.
• To enhance the exploration capability, convergence speed, and precision of the Gazelle Optimization Algorithm.

Main Methods:

• Developed MIGOA with population initialization using an optimal point set, tangent flight search, adaptive step size, and novel exploration strategies.
• Evaluated MIGOA on CEC2017 and CEC2020 benchmark test sets against GOA and eight other algorithms.
• Applied MIGOA to 3D UAV path planning and two engineering design problems.

Main Results:

• MIGOA demonstrated superior performance on CEC2017 and CEC2020 benchmark functions, achieving excellent average rankings.
• Statistical validation using Wilcoxon rank-sum and Friedman mean rank tests confirmed MIGOA's effectiveness.
• MIGOA successfully addressed constrained optimization problems in UAV path planning and engineering design.

Conclusions:

• The proposed MIGOA significantly improves upon the standard GOA, offering enhanced exploration, faster convergence, and higher precision.
• MIGOA shows strong scalability and practical applicability for solving complex, real-world optimization challenges.
• The enhanced algorithm effectively avoids local optima, making it a robust tool for various optimization tasks.