Rank-driven salp swarm algorithm with orthogonal opposition-based learning for global optimization.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Swarm Intelligence

Background:

• The Salp Swarm Algorithm (SSA) is a meta-heuristic optimization algorithm inspired by salp behavior.
• Standard SSA exhibits limitations in balancing exploration and exploitation, leading to slow convergence.
• There is a need for improved SSA variants to tackle complex optimization challenges effectively.

Purpose of the Study:

• To propose an enhanced Salp Swarm Algorithm (OOSSA) with improved comprehensive performance.
• To overcome the limitations of the basic SSA, including unbalanced operations and slow convergence.
• To validate the efficacy of OOSSA on mathematical, engineering, and real-world problems.

Main Methods:

• Developed an orthogonal lens opposition-based learning technique to escape local optima.
• Implemented adaptive adjustment of the number of leaders to enhance global exploration and convergence speed.
• Applied a dynamic learning strategy to improve the exploitation capability of the algorithm.

Main Results:

• OOSSA demonstrated superior performance over standard SSA and other state-of-the-art algorithms on 26 mathematical functions.
• Validated through Wilcoxon signed-rank and Friedman statistical tests.
• Successfully applied OOSSA to engineering optimization problems, photovoltaic model parameter extraction, and autonomous mobile robot path planning, generating optimal collision-free trajectories.

Conclusions:

• The proposed OOSSA significantly enhances the performance of the basic SSA.
• OOSSA effectively solves numerical, engineering, and real-world optimization problems, including path planning.
• The developed algorithm offers a robust and efficient solution for various complex optimization tasks.