摘要
用于工业分拣领域的机械装置通常是针对特定应用场景和特定产品而设计的,面对多种物品无序堆叠的场景,其普适性和智能性往往较差。当前基于3D结构光相机的点云匹配抓取技术虽在一定程度上提升了柔性生产能力,但受限于硬件成本高昂,以及特征描述能力有限、计算复杂度高、对遮挡敏感等固有缺陷,难以满足无序混装抓取需求。近年来以GraspNet为代表的深度学习抓取技术发展迅速,通过双目相机实现位姿估计,但仍存在目标选择策略欠优、位姿评分机制具有局限性、位姿定位偏差大等问题。针对上述挑战,提出一种改进型三阶段抓取算法。第一阶段,针对目标选择策略欠佳的问题,通过融合YOLOv10目标检测与SAM分割模型,结合优化的目标选择算法,即选择无遮挡、距离近的目标,有效解决了堆叠遮挡场景下的目标选择策略不佳难题。第二阶段,对GraspNet位姿估计框架进行改进,即通过引入基于点云表面法向量的位姿筛选机制,重构更加合理的评分机制,进而获取高精度抓取位姿。第三阶段,设计位姿微调策略,即采用"悬停对齐-垂直抓取"的分层控制架构,最大程度消除执行过程中的累积误差,有效解决位姿定位偏差大、实际抓取不准确问题。实验结果表明,该方法显著提升了复杂场景下的抓取效率、操作可靠性和跨场景泛化能力,同时由于使用双目相机取代了3D结构光相机,还显著降低了系统成本,为工业自动化提供了高性价比的解决方案。
Mechanical devices used in industrial sorting are typically designed for specific application scenarios and products,often exhibiting poor versatility and intelligence when faced with unordered mixed object grasping.Current point cloud matching grasping technologies based on 3D structured light cameras have improved flexible production capabilities to a certain extent.How-ever,they are constrained by high hardware costs,limited feature description capabilities,high computational complexity,and sensitivity to occlusions,making it difficult to meet the demands of unordered mixed object grasping.In recent years,deep learning-based grasping technologies,represented by GraspNet,have developed rapidly,achieving pose estimation through binocular ca-meras.Nevertheless,these methods still suffer from suboptimal target selection strategies,limitations in pose scoring mechanisms,and significant pose localization errors.To address these challenges,this study proposes an improved three-stage grasping algorithm.In the first stage,the YOLOv10 object detection model is fused with the SAM segmentation model,combined with an optimized target selection algorithm that prioritizes unobstructed and closer targets,effectively solving the problem of poor target selection strategies in stacked and occluded scenarios.In the second stage,the GraspNet pose estimation framework is enhanced by introducing a pose filtering mechanism based on point cloud surface normals and reconstructing the scoring mechanism to obtain high-precision grasping poses.In the third stage,a pose fine-tuning strategy is designed using a hierarchical control architecture of“hover alignment-vertical grasping”to effectively eliminate cumulative errors during execution,ultimately addressing the issue of inaccurate real-world grasping.Experimental results demonstrate that this method significantly improves grasping efficiency,operational reliability,and cross-scenario generalization capabilities in complex environments.Moreover,by replacing 3D structured light cameras with binocular cameras,the system cost is significantly reduced,providing a cost-effective solution for industrial automation.
作者
于灵鑫
陈艺博
曲浩君
厉广伟
李金屏
YU Lingxin;CHEN Yibo;QU Haojun;LI Guangwei;LI Jinping(School of Information Science and Engineering,University of Jinan,Jinan 250022,China;Shandong Provincial Key Laboratory of Network Based Intelligent Computing(University of Jinan),Jinan 250022,China;Shandong College and University Key Laboratory of Information Processing and Cognitive Computing in 13th Five-year(University of Jinan),Jinan 250022,China)
出处
《计算机科学》
北大核心
2026年第4期318-325,共8页
Computer Science
基金
山东省科技型中小企业创新能力提升工程(2022TSGC1047)
中央引导地方科技发展项目(YDZX2024078)
济南大学2023年学科交叉会聚建设项目(XKJC-202310)。
关键词
无序混装抓取
位姿估计
目标选择
姿态优化
双目相机
Unordered mixed objects grasping
Pose estimation
Target selection
Pose optimization
Binocular camera
