| 64 | 0 | 30 |
| 下载次数 | 被引频次 | 阅读次数 |
为解决乒乓球捡球机器人设计中的球体定位问题，提出一种基于图像与深度数据融合的方法。首先利用2D激光雷达与地面平行的特征，采用棋盘格平面标定板对激光雷达进行外参标定；再利用YOLOv8深度学习模型通过相机采集的图像检测出乒乓球的近似位置，据此对激光雷达获取到的球体扫描点进行筛选；最后利用Hough变换由筛选后的扫描点求出乒乓球球心坐标。真实实验表明，采用文中方法对乒乓球进行定位的平均误差为3.0mm，最大误差5.5mm，可满足捡球机器人对乒乓球的定位精度要求。
Abstract：A method based on image and depth data fusion was proposed to address the problem of ball localization in ping pong ball picking robot design. Firstly a chessboard plane calibration pattern was used to calibrate the extrinsic parameters of 2D LiDAR according to the parallel property between the LiDAR scanning plane and the ground. Secondly the approximate position of the ping pong ball was detected from the camera image using YOLOv8 deep learning model. The scanning points of the ping pong ball were selected from depth data from 2D LiDAR. At last the coordinates of the center of the ball was obtained by Hough transform using the filtered scanning points.Real experiments showed applying the proposed method, the average positioning error of the ping pong ball was 3.0mm, and the maximum error was 5.5mm, which satisfied the accuracy requirement of the ball picking robot.
[1]高长旭,孙巧妍,陈祥光,等.基于单目视觉的全自动捡球机器人控制算法设计[J].传感器与微系统, 2022, 41(09):88-91.
[2]刘冀龙,尹岗,赵建伟.采用图像处理技术的机械臂定位装置[J].电子设计工程, 2017, 25(21):161-163+168.
[3]周旭,周海波,陈睿,等.基于双目视觉的机械手捡球机器人设计[J].机床与液压, 2020, 48(03):34-40+79.
[4]戴亮亮.基于视觉的乒乓球及击球人识别系统研究[D].安徽:安徽工业大学, 2018.
[5]张涛.基于立体视觉的乒乓球运动轨迹检测方法研究[D].西安:西安理工大学,2024.
[6]孟群康,杨凯宁,李井源,等.基于FPGA的全自动定位捡乒乓球机器人系统设计[J].电子设计工程,2023,31(12):28-32+37.
[7]任云青.智能乒乓球自动捡球机器人的设计与实现[D].南京:南京邮电大学, 2020.
[8]李树青,邢鑫龙,林靖宇.一种三维激光雷达系统标定方法[J].激光与红外, 2022, 52(2):182-187.
[9] YUAN C X, CAI H B. Key nodes modeling for object detection and location on construction site using color-depth cameras[C]. 2014 International Conference on Computing in Civil and Building Engineering.ASCE. 2014:729-736.
[10] GENG H, GAO Z Y, FANG G R, et al. 3D object recognition and localization with a dense LiDAR scanner[J]. Actuators. MDPI, 2022, 11(1):13.
[11]江励,熊达明,汤健华,等.自然光线环境中的空间物体快速识别和定位算法研究[J].机械与电子, 2022, 40(06):8-13+21.
[12] ITAMI F, YAMAZAKI T. An improved method for the calibration of a 2D LiDAR with respect to a camera by using a checkerboard target[J].IEEE Sensors Journal, 2020, 20(14):7906-7917.
[13] VASCONCELOS F, BARRETO J, NUNES U. A minimal solution for the extrinsic calibration of a camera and a laser-rangefinder[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11):2097-2107.
[14]彭梦,邬书跃,陈龙,等.一种鲁棒的2D激光雷达和摄像机最小解标定方法[J].电子测量与仪器学报,2023,37(08):146-154.
[15] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified, real-time object detection[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA:IEEE, 2016:779-788.
基本信息:
DOI：
中图分类号:TP391.41;TP242
引用信息:
[1]李巍岷,罗寅,孙丽梅,等.基于图像与深度数据融合的球体定位[J].通信与信息技术,2025,No.277(05):16-21.
基金信息:
辽宁省教育厅高等学校基本科研重点项目（项目编号：LJKZ0243）; 辽宁省档案科技项目（项目编号：2023-X-050）