An Investigation of 6-DOF Robot Path Planning Using Evolutionary Algorithms

Multi-objective optimization (MOO) plays a vital role in robotics, where several conflicting objectives—such as minimizing energy consumption, execution time, and motion jerk—must be simultaneously satisfied. Traditional single-objective approaches fail to represent the trade-offs among such competing criteria, motivating the adoption of population-based evolutionary and swarm intelligence algorithms. This study presents a novel comparative investigation of two leading multi-objective optimization methods, MOPSO and NSGA-II, for 6-DOF robotic trajectory planning. The study also lays the foundation for a hybrid framework that sequentially combines both algorithms to leverage their complementary strengths. Both algorithms are applied to the multi-objective trajectory optimization of the AR4 robotic manipulator, aiming to achieve optimal motion performance under kinematic and dynamic constraints. The study evaluates each algorithm’s convergence behavior, diversity preservation, and computational efficiency using standard performance metrics—Generational Distance (GD), Spread (SP), and Hypervolume (HV). Results demonstrate that MOPSO exhibits rapid convergence and superior exploration capabilities, while NSGA-II achieves finer convergence accuracy and better Pareto front uniformity. Building on this complementarity, the paper proposes a future hybrid framework combining MOPSO’s exploratory strength with NSGA-II’s refinement and stability. The envisioned MOPSO→NSGA-II hybrid approach seeks to balance exploration and exploitation dynamically, achieving dense, smooth, and well-converged Pareto fronts. The findings provide a foundation for developing robust hybrid metaheuristics in robotic trajectory optimization, enhancing performance and adaptability in complex multi-objective environments.