基于灰狼优化的机器人视觉伺服协同控制  

Robotic Visual Servo Cooperative Control Based on Grey Wolf Optimizer

作　　者：牟雪琪 于海生[1,2] 张鹏鑫 杨庆 孟祥祥 MOU Xueqi;YU Haisheng;ZHANG Pengxin;YANG Qing;MENG Xiangxiang(School of Automation,Qingdao University,Qingdao 266071,China;Shandong Province Key Loratory of Industrial Control Technology,Qingdao University,Qingdao 266071,China)

机构地区：[1]青岛大学自动化学院,青岛266071 [2]青岛大学山东省工业控制技术重点实验室,青岛266071

出　　处：《组合机床与自动化加工技术》2025年第2期120-125,130,共7页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(62273189);山东省自然科学基金项目(ZR2021MF005)。

摘　　要：单独一种无标定视觉伺服控制难以使机器人末端实现快速和精确的目标定位,针对这一问题,提出了一种基于灰狼优化Kalman滤波算法视觉伺服与比例积分(proportional integral,PI)调节位置伺服的机器人协同控制策略。采用灰狼优化算法对Kalman滤波算法中的参数进行优化,提高了图像雅可比矩阵的在线估计精度,并设计了视觉伺服控制器。采用PI调节方法设计位置伺服控制器,提升了系统快速的动态性能。采用基于图像特征误差的高斯函数作为协同函数,设计了协同控制策略,兼顾了系统快速的动态性能和精确的稳态性能。仿真和实验结果表明,所提出的协同控制方法实现了更快速和更精确的目标定位。A single uncalibrated visual servo control faces challenges in achieving fast and accurate target position at the robot end-effector.To solve this issue,a visual servo based on grey wolf optimizer and PI-regulated position control cooperative control strategy for robots is proposed.Grey wolf optimizer was introduced to optimize the filtering parameters within the Kalman Filtering algorithm,enhancing the online estimation accuracy of the image Jacobian matrix.The visual servo controller was designed according to the above estimation.The proportional integral method was utilized in designing the position servo controller,which improved the system′s fast dynamic performance.The Gaussian function based on the visual image feature error was adopted as the cooperative function.The cooperative control strategy was designed to balance system fast dynamic performance with accurate steady-state performance.Simulation and experimental results indicate that the proposed method achieves faster and more accurate target location.

关 键 词：卡尔曼滤波 灰狼优化算法 无标定视觉伺服 协同控制 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]