MEIAO: A Multi-Strategy Enhanced Information Acquisition Optimizer for Global Optimization and UAV Path Planning.

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Optimization Algorithms

Background:

• Traditional path planning struggles with multi-constraint, multi-objective UAV missions in complex 3D environments.
• Standard swarm intelligence algorithms, like the Information Acquisition Optimizer (IAO), face limitations in exploration efficiency, population diversity, and boundary handling for high-dimensional problems.

Purpose of the Study:

• To develop an improved optimization algorithm, the Multi-Strategy Enhanced Information Acquisition Optimizer (MEIAO), for complex UAV path planning.
• To address the limitations of existing algorithms in exploration, diversity, and boundary handling within high-dimensional search spaces.

Main Methods:

• Incorporation of a Levy Flight-based strategy for enhanced global exploration.
• Integration of an adaptive differential evolution operator to balance exploration and exploitation dynamically.
• Implementation of a globally guided boundary handling strategy to ensure feasible path generation.
• Performance evaluation on CEC2017 and CEC2022 benchmark suites against eight other algorithms.

Main Results:

• MEIAO demonstrated superior performance in local exploitation, global exploration, and complex adaptation on benchmark functions.
• The algorithm exhibited enhanced robustness and maintained population diversity.
• Applied to 3D mountainous UAV path planning, MEIAO achieved a 25.7% reduction in average path cost compared to IAO.
• Generated paths were smoother, collision-free, and converged faster.

Conclusions:

• MEIAO provides a more efficient and reliable solution for unmanned aerial vehicle operations in complex 3D environments.
• The proposed algorithm effectively overcomes the limitations of existing methods in UAV path planning.
• MEIAO offers significant improvements in path cost, smoothness, and convergence speed.