Graduate Student Evolutionary Algorithm: A Novel Metaheuristic Algorithm for 3D UAV and Robot Path Planning.

Area of Science:

• Artificial Intelligence
• Optimization Algorithms
• Robotics

Background:

• Numerical optimization, UAVs, and robot path planning are critical AI research areas.
• Traditional methods face challenges with complex nonlinear problems, exhibiting slow convergence, low accuracy, and poor robustness.
• Intelligent optimization algorithms are needed to overcome limitations of existing approaches.

Purpose of the Study:

• To introduce a novel intelligent optimization algorithm, the Graduate Student Evolutionary Algorithm (GSEA).
• To address the limitations of current optimization techniques in terms of convergence speed, accuracy, and robustness.
• To evaluate GSEA's performance on benchmark test sets and its applicability to real-world path planning problems.

Main Methods:

• Developed a mathematical model for GSEA inspired by graduate student behaviors like research direction searching and focused study.
• Qualitatively analyzed the convergence behavior of GSEA.
• Evaluated GSEA's performance against competitive algorithms using the CEC2017 and CEC2022 test suites.
• Applied GSEA to Unmanned Aerial Vehicle (UAV) and robot path planning challenges.

Main Results:

• GSEA demonstrated superior performance compared to existing algorithms on CEC2017 and CEC2022 benchmark test sets.
• Statistical tests confirmed the effectiveness and robustness of GSEA.
• Experimental results showed GSEA's superiority in solving practical UAV and robot path planning problems.

Conclusions:

• GSEA is an effective and robust intelligent optimization algorithm.
• The proposed algorithm offers a promising solution for complex nonlinear optimization tasks.
• GSEA shows significant potential for practical applications in UAV and robot path planning.