A Billion Ways to Grasp

Name: A Billion Ways to Grasp
Keywords: research-lab

A Billion Ways to Grasp is a grasping robotics dataset maintained by NVIDIA et al.. Primary hardware platform: simulated parallel-jaw grasps, dense SE(3) sampling. Access is not stated under not stated.

Dataset specifications
Year	2019
Embodiments	simulated parallel-jaw grasps, dense SE(3) sampling
Task categories	manipulation
License	not stated
Access	open
Maintainer	NVIDIA et al.