Solving the 3D UAV Path Planning Problem Using an Improved Multi-Leader Multi-Objective Whale Optimization Algorithm

Area of Science:

• Robotics
• Artificial Intelligence
• Operations Research

Background:

• Unmanned Aerial Vehicle (UAV) path planning in complex 3D environments presents significant challenges due to conflicting objectives and constraints.
• Existing multi-objective whale optimization algorithms (MOWOA) often struggle with low feasible solution rates, unstable convergence, and poor Pareto solution distribution.

Purpose of the Study:

• To address limitations of MOWOA in constrained UAV path planning.
• To propose an improved multi-leader multi-objective whale optimization algorithm (IML-MOWOA) for enhanced path planning performance.

Main Methods:

• Formulated UAV path planning as a multi-objective optimization problem considering path length, threat, smoothness, and altitude costs.
• Introduced an adaptive opposition-based learning for initial population construction.
• Implemented a grid-based external archive strategy for solution density regulation.
• Developed a multi-leader dynamic weighted search mechanism with Softmax-based cosine annealing.

Main Results:

• IML-MOWOA demonstrated more robust convergence and better Pareto-front distribution compared to benchmark algorithms.
• Achieved a higher number of feasible paths in challenging scenarios.
• Significantly reduced mean Inverted Generational Distance (IGD) by 25.04%.
• Reduced mean path length, threat cost, smoothness cost, and altitude cost by 1.65%, 28.45%, 53.23%, and 29.88%, respectively.

Conclusions:

• The proposed IML-MOWOA is effective and robust for constrained multi-objective UAV path planning in complex static 3D environments.
• The enhancements improve solution quality, feasibility, and distribution, outperforming existing methods.