Multi-strategy Sea Horse Optimization algorithm for UAV path planning

Area of Science:

• Robotics and Autonomous Systems
• Artificial Intelligence
• Optimization Algorithms

Background:

• Unmanned aerial vehicle (UAV) path planning is crucial for autonomous navigation but presents complex optimization challenges.
• Traditional and metaheuristic methods often struggle with constraints and premature convergence to local optima.

Purpose of the Study:

• To introduce a modified Sea Horse Optimization (moSHO) algorithm for effective threat-aware UAV path planning.
• To enhance the exploration and exploitation balance in optimization for complex navigation tasks.

Main Methods:

• Developed moSHO by integrating three cooperative strategies into the Sea Horse Optimization algorithm: fish-aggregating device (FAD) mechanism, best-worst position mutation (BWPM), and quasi-reflection-based learning (QRBL).
• FAD promotes diversity via adaptive perturbations.
• BWPM refines elite solutions and guides weaker ones.
• QRBL enhances exploration using quasi-opposite candidates.

Main Results:

• The moSHO algorithm demonstrated robustness across 23 benchmark functions.
• Experiments confirmed moSHO's reliability in identifying safe and feasible UAV paths within threat environments.
• The integrated strategies improved exploration without sacrificing exploitation, leading to a balanced optimization process.

Conclusions:

• The proposed moSHO algorithm offers a superior approach to UAV path planning in complex, threat-aware scenarios.
• moSHO effectively overcomes limitations of traditional methods and enhances autonomous navigation capabilities.