| 摘要: |
| 目的 为了解决全自动食品包装机的块状食品抓取问题，以拾放并联机械手为研究对象，结合视觉反馈和预测控制，提出一种机械手定位控制方法。方法 给出了视觉反馈机械手总体设计方案，包括机械手本体、双目视觉系统和控制系统。研究输送带上运动食品的机器视觉定位算法。在考虑目标可见性约束和机械手执行器约束的前提下，将预测控制算法引入机械手的视觉伺服控制中，用于实现机械手末端位姿控制。最后，通过实验验证机械手的定位精度。结果 结果表明，平均误差可控制在0.7%，符合运动拾取精度要求。结论 所述控制方法可满足包装工艺要求，提高了包装过程自动化程度。 |
| 关键词: 包装机械手 机器视觉 预测控制 定位 |
| DOI：10.19554/j.cnki.1001-3563.2019.17.030 |
| 分类号:TB486 |
| 基金项目:江苏省青蓝工程资助项目（2017）；常州市科技支撑计划（工业）项目（CE20170069） |
|
| Packaging Manipulator Positioning Control Based on Visual Feedback |
|
YUAN Fei1, ZHOU Wen-bin2
|
|
1.Changzhou Vocational Institute of Light Industry, Changzhou 213000, China;2.Philips Healthcare (Suzhou) Co., Ltd., Suzhou 215024, China
|
| Abstract: |
| The work aims to propose a manipulator positioning control method by taking the pick-and-place parallel manipulator as the research object and combining visual feedback with predictive control, in order to solve the problem of bulk food grabbing of the automatic food packaging machine. The overall design scheme of the visual feedback manipulator was given, including the manipulator body, binocular vision system and control system. The machine vision localization algorithm of moving food on conveyor belt was studied. The predictive control algorithm was introduced into the visual servo control of the manipulator on the premise of considering the visual constraints of the target and the constraints of the manipulator effector. Finally, the positioning accuracy of the manipulator was verified by experiments. The results showed that, the average error could be controlled at 0.7%, which met the requirements of motion pickup accuracy. The proposed control method can meet the requirements of packaging technology and improve the automation degree of packaging process. |
| Key words: packaging manipulator machine vision predictive control positioning |
渝公网安备 50010702501716号