Robotic grasping is a fundamental skill required for object manipulation in robotics. Multi-fingered robotic hands, which mimic the structure of the human hand, can potentially perform complex object manipulation. Nevertheless, current techniques for multi-fingered robotic grasping frequently predict only a single grasp for each inference time, limiting computational efficiency and their versatility, i.e. unimodal grasp distribution. This paper proposes a differentiable multi-fingered grasp generation network (DMFC-GraspNet) with three main contributions to address this challenge. Firstly, a novel neural grasp planner is proposed, which predicts a new grasp representation to enable versatile and dense grasp predictions. Secondly, a scene creation and label mapping method is developed for dense labeling of multi-fingered robotic hands, which allows a dense association of ground truth grasps. Thirdly, we propose to train DMFC-GraspNet end-to-end using using a forward-backward automatic differentiation approach with both a supervised loss and a differentiable collision loss and a generalized Q 1 grasp metric loss. The proposed approach is evaluated using the Shadow Dexterous Hand on Mujoco simulation and ablated by different choices of loss functions. The results demonstrate the effectiveness of the proposed approach in predicting versatile and dense grasps, and in advancing the field of multi-fingered robotic grasping.

翻译：机器人抓取是物体操作中的一项基础技能。多指机器人手通过模仿人手结构，有潜力执行复杂的物体操作任务。然而，当前的多指机器人抓取技术通常在每次推理时仅预测单一抓取位姿，这限制了计算效率与多功能性（即单模态抓取分布）。本文提出一种可微分多指抓取生成网络（DMFC-GraspNet），通过三项主要贡献应对这一挑战。首先，提出一种新型神经抓取规划器，可预测一种全新的抓取表示，以实现多功能且密集的抓取预测。其次，开发了一种场景创建与标签映射方法，用于多指机器人手的密集标注，从而建立真实抓取位姿的密集关联。第三，我们提出采用前向-后向自动微分方法，结合监督损失、可微分碰撞损失及广义Q¹抓取度量损失，对DMFC-GraspNet进行端到端训练。该方法在Mujoco仿真环境中使用Shadow Dexterous Hand进行评估，并通过不同损失函数的选择进行消融实验。结果表明，该方法在预测多功能密集抓取位姿方面具有有效性，并推动了多指机器人抓取领域的发展。