Sim-MEES: Modular End-Effector System Grasping Dataset for Mobile Manipulators in Cluttered Environments

This is the sample dataset part for our paper: Sim-MEES

🚀🚀 New Work 🚀🚀 Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark (https://junchengli1.github.io/Sim-Suction/)

seg_dic - object segmentation dictionary.

mass_dic - object mass dictionary.

difficulty_dic - object difficulty level dictionary.

/Stage_0:

/depth - (800 frames) depth images

/bbox_3d - (800 frames) 3d bounding box

/bbox_2d_tight - (800 frames) 2d bounding box

/instance - (800 frames) segmentation mask

/rgb - (800 frames) RGB image

/camera_info - Camera matrix

0.npz - point clouds, surface normals, and segmentation ID for each point.

/.*pcd - point cloud of each objects in the cluttered enviroments.

stage_0.usd - cluttered environments usd file.

Examples stage grasps visualization

We provide visualization tool to see example suction and parallel jaw gripper grasp labels in folder:

Prerequisite: open3d

Visualize stage and suction grasp labels: view_suction.py

Visualize stage and parallel jaw grasp labels: view_parallel_jaw.py

label details

(1.5 cm suction cup gripper) stage_0_candidates_after_seal.pkl

["segmentation_id"]
["object_name"]
["rotation_after_seal_pass"]
["translation_after_seal_pass"]
["rotation_after_seal_fail"]
["translation_after_seal_fail"]
["translation_after_collision_pass"]
["rotation_after_collision_pass"]
["translation_after_collision_fail"]
["rotation_after_collision_fail"]
["total_candidates"] 
["total_candidates_pass_collision"]
["total_candidates_pass_seal"]

(1.5 cm suction cup gripper) stage_0_seal_simulation_candidates.pkl

["rotation_after_exp_success"]
["translation_after_exp_success"]
["rotation_after_exp_fail"]
["translation__after_exp_fail"]
["object_mass"]

(2.5 cm suction cup gripper) stage_0_candidates_after_seal_2.5cm.pkl

["segmentation_id"]
["object_name"]
["rotation_after_seal_pass"]
["translation_after_seal_pass"]
["rotation_after_seal_fail"]
["translation_after_seal_fail"]
["translation_after_collision_pass"]
["rotation_after_collision_pass"]
["translation_after_collision_fail"]
["rotation_after_collision_fail"]
["total_candidates"] 
["total_candidates_pass_collision"]
["total_candidates_pass_seal"]

(2.5 cm suction cup gripper) stage_0_seal_simulation_candidates_2.5cm.pkl

["rotation_after_exp_success"]
["translation_after_exp_success"]
["rotation_after_exp_fail"]
["translation__after_exp_fail"]
["object_mass"]

(Fetch) stage_0_grasp_candidates_after_overlap.pkl

["segmentation_id"]
["object_name"]
["rotation_after_overlap_pass"]
["translation_after_overlap_pass"]
["rotation_after_overlap_fail"]
["translation_after_overlap_fail"]
["total_candidates"]
["total_candidates_pass_overlap"] 

(Fetch) stage_0_grasp_simulation_candidates.pkl

["rotation_after_exp_success"]
["translation_after_exp_success"]
["rotation_after_exp_fail"]
["translation__after_exp_fail"]