融合学习与几何优化，提升部分观测下的抓取成功率。
A Hybrid Optimization Framework for Grasp Synthesis under Partial Observations
- 用能量模型引导变分梯度迭代优化抓取姿态
- 在5360次尝试中平均成功率达60.9%
- 适合需要鲁棒抓取的机器人操作场景
我们提出一种混合抓取合成框架，结合基于学习的能量模型（EBM）与解析的迭代最近点（ICP）方法，从部分观测点云中生成鲁棒抓取。学习到的能量函数作为先验，在斯坦因变分梯度下降（SVGD）框架内引导抓取配置的迭代优化。在67个物体上进行5360次抓取尝试的评估显示，该方法平均成功率为60.9%，优于AnyGrasp（31.1%）、Grasp Pose Detection（48.4%）和AS-ICP（56.6%）。结果表明该方法具备强泛化能力，展示了数据驱动学习与几何优化结合如何克服单一策略的局限性。
原文摘要 · Abstract (English)
We propose a hybrid grasp synthesis framework that combines a learning-based Energy-Based Model (EBM) with an analytical Iterative Closest Point (ICP) method to generate robust grasps from partially observed point clouds. The learned energy function acts as a prior within a Stein Variational Gradient Descent (SVGD) framework, guiding iterative refinement of grasp configurations. Evaluated on 67 objects with 5,360 grasp attempts, our method achieves an average success rate of 60.9\%, outperforming AnyGrasp (31.1\%) and Grasp Pose Detection (48.4\%) and AS-ICP (56.6\%). These results highlight the strong generalization ability of our approach and demonstrate how combining data-driven learning with geometric optimization addresses the limitations of either strategy in isolation.
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。