2D and 3D Grasp Planners for the GET Asymmetrical Gripper

2026-04-28

AI Summary

This work addresses the challenge of efficient and stable grasp planning for the GET asymmetric gripper from single-view RGB-D images. We propose two novel approaches: GET-2D-1.0, a fast 2D planner leveraging the Ferrari–Canny metric with a tailored sampling strategy, and GET-3D-1.0, a voxelized 3D planner incorporating a full 3D gripper model and ray tracing. As the first grasp planning framework specifically designed for the GET gripper, our method introduces an efficient sampling scheme adapted to single-view inputs and provides a systematic comparison between 2D and 3D formulations. Experiments demonstrate that GET-2D-1.0 improves grasp success rate, robustness to perturbations, and force resistance by over 40% compared to a bounding-box baseline, achieving planning in just 683 ms; GET-3D-1.0 yields marginal performance gains at the cost of approximately 17 seconds per plan.

Abstract

In this paper, we introduce GET-2D-1.0, a fast grasp planner for the GET asymmetrical gripper that operates from a single-view RGB-D image, using the Ferrari-Canny metric and a novel sampling strategy, and GET-3D-1.0, a mesh-based method using a 3D gripper model and ray-tracing. We evaluate both grasp planners against baselines with physical experiments, which suggest that GET-2D-1.0 can improve over a bounding box baseline by over 40% in lift success, shake survival, and force resistance. Experiments with GET-3D-1.0 suggest slight improvement compared to GET-2D-1.0 on lift success and shake survival, but are more computationally expensive, averaging 17 seconds of planning compared to 683 ms for GET-2D-1.0.

Problem

• grasp planning
• asymmetrical gripper
• RGB-D image
• 3D mesh
• robotic grasping

Innovation

• asymmetrical gripper
• grasp planning
• Ferrari-Canny metric
• ray-tracing
• RGB-D-based grasping