针对细小脆弱物体的抓取难题，提出改进方法提升成功率。
FineGrasp: Towards Robust Grasping for Delicate Objects
- 通过网络结构优化增强对精细区域的感知能力
- 设计新标签归一化策略缓解标注不平衡问题
- 构建新仿真数据集并结合真实场景训练提升泛化性
近年来机器人抓取技术已广泛集成到各类操作系统中，例如基于语言的语义分割可实现对任意指定物体或部位的抓取。然而，现有方法在生成小物体或精细部件的可行抓取姿态时仍存在困难，可能导致整个流程失败。为此，我们提出一种新型抓取方法 FineGrasp，从三个关键方面进行改进：第一，引入多种网络结构调整以增强对精细区域的处理能力；第二，解决标签不平衡问题，提出优化的抓取度标签归一化策略；第三，构建新的模拟抓取数据集，并验证混合仿真实-真实训练可进一步提升抓取性能。实验结果表明，该方法在小物体抓取任务中取得显著提升，证实了其在语义抓取中的有效性。
原文摘要 · Abstract (English)
Recent advancements in robotic grasping have led to its integration as a core module in many manipulation systems. For instance, language-driven semantic segmentation enables the grasping of any designated object or object part. However, existing methods often struggle to generate feasible grasp poses for small objects or delicate components, potentially causing the entire pipeline to fail. To address this issue, we propose a novel grasping method, FineGrasp, which introduces improvements in three key aspects. First, we introduce multiple network modifications to enhance the ability of to handle delicate regions. Second, we address the issue of label imbalance and propose a refined graspness label normalization strategy. Third, we introduce a new simulated grasp dataset and show that mixed sim-to-real training further improves grasp performance. Experimental results show significant improvements, especially in grasping small objects, and confirm the effectiveness of our system in semantic grasping.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。