首页|基于深度学习的产品分拣机器人设计

基于深度学习的产品分拣机器人设计

扫码查看
随着人工智能时代的到来,智能机器人被广泛运用在各个领域中,其中分拣式机器人的运用最为广泛,由于其高效快速地完成现代生产生活中的分类工作,因此在工业生产和环境科学等领域都有其身影.通过分析快递行业分拣工作存在的问题,提出了用分拣机器人进行分拣工作的策略,利用Tensorflow框架建立CNN神经网络模型,通过Keras-YOLOV3算法实现将物体归类并生成带有分类以及置信度的锚框,设计了基于深度学习的产品分拣机器人,实现对工厂流水线零件的分类和通过摄像头识别工厂流水线生产零件并利用机械臂完成零件分类工作.
Design of product sorting robot based on deep learning
With the arrival of the era of artificial intelligence,intelligent robots are widely used in various fields,among which sorting robots are the most widely used,due to their efficient and rapid completion of the sorting work in modern production and life,so they are found in industrial production and environmental science and other fields.By analyzing the problems of sorting work in the courier industry,the strategy of sorting work with sorting robots is proposed,the CNN neural network model is established using the Tensorflow framework,the Keras-YOLOV3 algorithm is used to realize the categorization of the objects and to generate the anchor frames with the classification as well as the confidence level,and the product sorting robot based on deep learning is designed to realize the classification of the parts of the factory line and the recognition of the parts of the factory line through the camera.The deep learning based product sorting robot is designed to classify the parts of the factory assembly line and to recognize the production parts of the fac-tory assembly line through the camera and to complete the parts sorting work by using the robotic arm.

Robotic arm sortingRobotKeras-YOLO3 algorithmDeep learning

赵洁、薛若银、刘玉升

展开 >

天津城建大学计算机与信息工程学院,天津 300384

机械臂分拣 机器人 Keras-YOLO3算法 深度学习

2024

通信与信息技术
四川省通信学会

通信与信息技术

影响因子:0.223
ISSN:1672-0164
年,卷(期):2024.(1)
  • 11