基于灰狼算法的球形机器人构件参数优化

Optimization of Spherical Robot Component Parameters Based on Grey Wolf Optimization Algorithm

摘要: 双摆驱动式球形机器人，以其零转弯半径、高实用性、灵活性和密封性著称。为提升性能并促进后续控制及路径研究，本研究对同轴双摆球形机器人进行了运动建模分析。通过受力分析设计代价函数，并运用灰狼优化算法(GWO)求解，与粒子群优化算法(PSO)对比，GWO在七次迭代后即收敛，表现更优。仿真结果显示，GWO优化的机器人球壳运动速度达2.681 m/s，优于PSO的2.513 m/s，验证了GWO的高效性及代价函数的有效性。这一成果为球形机器人运动系统设计的合理性提供了实证，对相关领域有参考意义。

Abstract: Spherical robots with dual pendulum drive are celebrated for their zero turning radius, exceptional practicality, agility, and tight sealing. To enhance performance and advance further studies in control and path planning, this research undertook a motion modeling analysis of a coaxial double pendulum spherical robot. A cost function was devised based on force analysis and solved using Grey Wolf Optimization (GWO), contrasted with the Particle Swarm Optimization (PSO) algorithm. GWO converged after just seven iterations, outperforming PSO. Simulation results revealed that the robot’s spherical shell speed optimized by GWO reached 2.681 m/s, surpassing PSO’s 2.513 m/s, thereby validating GWO’s efficiency and the efficacy of the cost function. This achievement substantiates the rationality of spherical robot motion system designs and offers valuable insights for related fields.