拼图机器人视觉控制系统设计与应用  

Design and Application of Vision Control System for Puzzle Robot

作　　者：陈晋生 赖惠鸽[1] 史伟 丁云鹏 CHEN Jin-sheng;LAI Hui-ge;SHI Wei;DING Yun-peng(School of Mechanical Engineering,Ningxia University,Yinchuan 750071,China)

机构地区：[1]宁夏大学机械工程学院,宁夏银川750021

出　　处：《控制工程》2021年第4期781-786,共6页Control Engineering of China

基　　金：国家自然科学基金资助项目(51765056)。

摘　　要：为研究拼图机器人视觉系统的控制稳定性,确保拼图机器人在拼图七巧板的过程中能够实现90%以上的准确率,以KR6型机器人为研究对象,依据空间坐标系变换和机械臂逆运动学解的理论基础,建立KR6型机器人视觉控制系统标定模型。利用Canny边缘检测算法和Douglas-Peucker算法对七巧板模型进行图像识别定位,并采用五次多项式插值法完成拼图轨迹规划。通过仿真模拟实验得出,KR6型机器人可以实现对七巧板的识别定位,并沿着预定的轨迹完成拼图任务。最后,通过在现场搭建实验平台,进一步验证了所设计的拼图机器人视觉控制系统能够较精准地完成各种预设图案的拼图。In order to study the control stability of the puzzle robot vision system and ensure that the puzzle robot can achieve more than 90% accuracy in the process of jigsaw puzzle, taking the KR6 robot as the research object and according to the theoretical basis of the space coordinate system transformation and the inverse kinematics solution of the manipulator, the calibration model of the robot vision control system is established. The Canny edge detection algorithm and the Douglas-Peucker algorithm are used for recognition and positioning of the tangram model, and the fifth degree polynomial interpolation method is used to complete the puzzle trajectory planning. Through the simulation experiments, it is concluded that the KR6 robot can realize the recognition and positioning of the jigsaw puzzle and complete the handling puzzle task along the predetermined trajectory. Finally, by setting up the experimental platform on site, it is further verified that the designed vision control system for puzzle robot can complete the puzzles of various preset patterns accurately.

关 键 词：机器人视觉系统 KR6型机器人 CANNY边缘检测算法 DOUGLAS-PEUCKER算法 轨迹规划 

分 类 号：TP249[自动化与计算机技术—检测技术与自动化装置]