Leveraging CVAE for Joint Configuration Estimation of Multifingered Grippers from Point Cloud Data

Abstract-- This paper presents an efficient approach for determining the joint configuration of a multifingered gripper solely from the point cloud data of its poly-articulated chain, as generated by visual sensors, simulations or even generative neural networks. Well-known inverse kinematics (IK) techniques can provide mathematically exact solutions (when they exist) for joint configuration determination based solely on the fingertip pose, but often require post-hoc decision-making by considering the positions of all intermediate phalanges in the gripper's fingers, or rely on algorithms to numerically approximate solutions for more complex kinematics. In contrast, our method leverages machine learning to implicitly overcome these challenges. This is achieved through a Conditional V ariational Auto-Encoder (CV AE), which takes point cloud data of key structural elements as input and reconstructs the corresponding joint configurations. This highlights the effectiveness of our pipeline for joint configuration estimation within the broader context of AI-driven techniques for grasp planning. Determining joint configurations for multi-fingered robotic grippers is a critical challenge from a control perspective, as precise joint control is essential for accurately positioning fingertips or phalanges at the desired contact points on the object. Indeed, several well-known approaches for generating valid grasps rely on analytical metrics, such as force-or form-closure criteria [1]-[3].