基于计算机视觉的机械手包裹分拣与放置系统

A Robot Parcel Sorting and Placement System Based on Computer Vision

物流业发展迅速,但快递工人的数量有限,快递员总是要处理成堆的包裹,工作效率低下。针对分拣和投递系统的快件堆放问题,提出并设计了一种基于视觉的机械手快递系统。该系统结合了计算机视觉、深度学习、机械手控制和运动规划技术。采用YOLO算法和PCL法向量确定位置和法向量包裹取件点,规划并控制机械手移动到目标位置吸包,并通过视觉确定包裹尺寸,获取客户信息。与RCNN和SDD算法相比,YOLO算法具有明显的收敛性,检测速度更快,同时对快递箱等小型堆叠目标的处理精度更高。采用YOLOv3算法来提高系统的提取精度和速度。最后,系统的成功率为75%,完成时间约为15 s。实验结果表明,该系统具有分拣和快递功能,以及向客户发送取货信息的功能,具有一定实际应用和借鉴价值。

The logistics industry is developing rapidly,but the number of courier workers is limited,and couriers always have to handle piles of packages,resulting in low work efficiency.A vision based robotic delivery system is proposed and designed to address the issue of stacking parcels in sorting and delivery systems.The system combines computer vision,deep learning,robotic arm control,and motion planning technologies.Using YOLO algorithm and PCL normal vector to determine the position and normal vector to wrap the pickup point,planning and controlling the robotic arm to move to the target position to suck the package,and visually determining the package size to obtain customer information.Compared with RCNN and SDD algorithms,YOLO algorithm has significant convergence,faster detection speed,and higher processing accuracy for small stacked targets such as express boxes.Using YOLOv3 algorithm to improve the extraction accuracy and speed of the system.Finally,the success rate of the system is 75%,and the completion time is approximately 15 s.The experimental results indicate that the system has sorting and express delivery functions,as well as the ability to send pick-up information to customers,which has certain practical application and reference value.