Time-optimal trajectory planning for a six-degree-of-freedom manipulator: a method integrating RRT and chaotic PSO

This study proposes an innovative algorithm based on a hybrid optimization strategy, integrating the rapidly-exploring random tree (RRT) and an improved particle swarm optimization (PSO) method to address the time-optimal trajectory planning problem for six-degree-of-freedom robotic arms. The proposed approach emphasizes obstacle avoidance and motion smoothness. RRT is utilized within a dynamic three-dimensional environment to rapidly generate an initial collision-free path. Subsequently, improved PSO enhances global search performance by introducing a multi-source chaotic mapping-based population initialization strategy, dynamically adjusted inertia weights, and nonlinear learning factors. These enhancements effectively mitigate the limitations of traditional PSO methods, which are prone to premature convergence in complex optimization problems. Furthermore, the proposed 3-5-3 polynomial interpolation method significantly smooths the trajectory, reducing fluctuations in velocity and acceleration, and thereby improving the precision and energy efficiency of trajectory planning. Experimental results demonstrate that the proposed algorithm outperforms existing methods, such as the improved RRT and improved non-dominated sorting genetic algorithm, across multiple metrics. Notable achievements include a reduction of total motion time by approximately 21%, improved stability in robotic arm motion, and enhanced adaptability to dynamic environments. Particularly, the method achieves superior trajectory diversity and uniformity through joint optimization of multiple chaotic sources, overcoming the inherent limitations of single chaotic mappings. This research expands the application scenarios of RRT and PSO and provides a novel solution for intelligent control and real-time planning of high-degree-of-freedom robotic arms.