基于视觉感知的机械臂控制系统研究
杜金平1,2,周燕1,2,朱春颖1,2
(1 河北机电职业技术学院 电气工程系,河北 邢台 054000;
2 邢台市智能工厂监控与网络技术创新中心,河北 邢台 054000)
摘 要 :为了解决传统机械臂在非结构化环境中适应性差、定位精度低、动态跟随能力不足等问题,设计了一种融合深度学习与视觉伺服的机械臂智能控制系统。该系统采用改进 YOLOv8 算法实现目标的快速检测与特征提取,结合手眼标定与立体视觉技术完成三维空间位置解算,并提出基于模糊 PID 的多轴协同控制策略,解决了机械臂运动过程中的抖动与超调问题。实验结果表明,该系统在静态目标识别准确率较高,响应时间缩短,相比传统控制系统综合性能得到了提升,为工业机器人的应用提供了技术支持。
关键词 : 视觉感知 ;机械臂 ;深度学习 ;手眼标定 ;模糊 PID ;视觉伺服
中图分类号 :TP241 文献标识码 :A 文章编号 :2097-6623(2026)10-0040-05
Research on Manipulator Control System
Based on Visual Perception
DU Jin-ping1, 2, ZHOU Yan1, 2, ZHU Chun-ying1, 2
( 1 Department of Elecerical Engineering, Hebei Mechanical and Electrical Technology Polytechnic, Xingtai 054000, China;
2 Xingtai Technology Innovation Center for Intelligent Factory Monitoring and Industrial Network, Xingtai 054000, China)
Abstract: To address the issues of poor adaptability, low positioning accuracy, and insufficient dynamic tracking ability of traditional manipulators in unstructured environments, an intelligent control system for manipulator integrating deep learning and visual servoing is designed. The system uses the improved YOLOv8 algorithm to achieve rapid detection and feature extraction of targets, combines hand-eye calibration and stereo vision technology to solve the three-dimensional spatial position calculation. Furthermore, a multi-axis coordinated control strategy based on fuzzy PID is proposed to suppress the jitter and overshoot problems during manipulator motion. Experimental results show that the system achieves high recognition accuracy for static targets and reduces response time. Compared with traditional control systems, its overall performance is improved, which provides technical support for the application of industrial robots.
Key words: visual perception; manipulator; deep learning; hand-eye calibration; fuzzy PID; visual servoing
参考文献
[1] 周凌云,关哲,华长春,等 . 基于增强型 PPO-PID 的机械臂跟踪控制 [J]. 控制与决策,2026,41(8) :2089-2096.
[2] 王银,杨洁 . 永磁同步电机驱动的机械臂自抗扰控制研究 [J]. 机械工程与自动化,2026,55(2) :160-163.
[3] 余顺奇,郭宇飞,许盛悦,等 . 基于物理信息神经网络的振动基机械臂鲁棒控制 [J]. 中国仪器仪表,2026(3) :27-32.
[4] 李秉霖,白宇文,赵常屹,等 . 一种基于抗饱和滑模面与自适应趋近率的机械臂轨迹跟踪控制方法 [J]. 制造业自动化,2026,48(3) :1-8.
[5] 朱栩,仝思佳 . 基于模糊补偿的水下机械臂自适应控制 [J]. 南方农机,2026,57(5) :146-149.
[6] 苏竹阳,陈锋,柳清 . 基于 MPC-PD 分层控制的机械臂高精度轨迹跟踪方法 [J]. 淮北师范大学学报 ( 自然科学版 ),2026,47(3) :55-61.
[7] 魏琼, 陈子超, 李旭东,等 . 三维气动软体机械臂静力学建模与控制方法研究 [PP/OL]. 机械设计(2026-03-11)[2026-04-13].https://link.cnki.net/doi/10.13841/j.cnki.jxsj.20260310.001.
[8] 陈乙铭,潘毅峰,张承谦,等 . 柔性机械臂系统的鲁棒运动控制算法设计与实验验证 [J]. 机电工程,2026,43(6) :1221-1229.
[9] 何佳容 . 工业机器人机械臂抓握运动自动化控制研究 [J].机械管理开发,2026,41(2) :18-20.
[10] 郑明理 . 机械臂自动化驱动与控制策略研究 [J]. 今日制造与升级,2026(2) :150-153.