A Dynamic Multi-Niche Biogeography-Based Optimization Algorithm and Its Application to Robot Path Planning.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• Biogeography-based optimization (BBO) is a population-based metaheuristic algorithm inspired by species migration.
• The original BBO algorithm faces challenges such as premature convergence and insufficient population diversity in complex optimization tasks.

Purpose of the Study:

• To propose a novel dynamic multi-niche biogeography-based optimization (DMBBO) algorithm to overcome the limitations of the standard BBO.
• To enhance population diversity, parallel search capability, information exchange efficiency, and solution quality.

Main Methods:

• Developed a dynamic multi-niche population structure for improved diversity and parallel search.
• Implemented a dual-source migration mechanism to enhance information exchange.
• Introduced a niche-level hybrid elite preservation strategy for stable convergence and better solution quality.
• Conducted extensive experiments on CEC2022, CEC2020, and CEC2019 benchmark test suites and applied the algorithm to robotic path planning problems.

Main Results:

• DMBBO consistently outperformed 23 state-of-the-art algorithms across various dimensions.
• Achieved statistically significant improvements in optimization accuracy, convergence speed, and robustness, validated by Friedman and Wilcoxon tests.
• Demonstrated effectiveness through ablation studies and convergence behavior analysis.
• Successfully generated shorter and more stable paths in robotic path planning across diverse scenarios.

Conclusions:

• The proposed DMBBO algorithm effectively addresses the limitations of traditional BBO.
• DMBBO exhibits superior performance and robustness compared to existing algorithms on benchmark test suites.
• The algorithm shows strong applicability and practical value for constrained optimization problems, such as robotic path planning.