The Grasp Reset Mechanism: an Automated Apparatus for Conducting Grasping Trials

Advancing robotic grasping and manipulation requires the ability to testalgorithms and/or train learning models on large numbers of grasps. Towards thegoal of more advanced grasping, we present the Grasp Reset Mechanism (GRM), afully automated apparatus for conducting large-scale grasping trials. The GRMautomates the process of resetting a grasping environment, repeatably placingan object in a fixed location and controllable 1-D orientation. It alsocollects data and swaps between multiple objects enabling robust datasetcollection with no human intervention. We also present a standardized statemachine interface for control, which allows for integration of mostmanipulators with minimal effort. In addition to the physical design andcorresponding software, we include a dataset of 1,020 grasps. The grasps werecreated with a Kinova Gen3 robot arm and Robotiq 2F-85 Adaptive Gripper toenable training of learning models and to demonstrate the capabilities of theGRM. The dataset includes ranges of grasps conducted across four objects and avariety of orientations. Manipulator states, object pose, video, and graspsuccess data are provided for every trial.