{"success":true,"data":{"author":"Artificial and Mechanical Intelligence (AMI) lab, Istituto Italiano di Tecnologia (curated by Giovanni Fregonese)","bucketAccessType":"public","bucketName":"ami-iit/puzzle_manipulation_datasets","bucketPrefix":"","category":"Robotics","cloudProvider":"huggingface","createdAt":"2026-11-21T22:15:33.035Z","description":"Logged teleoperation data from ergoCub, the humanoid robot of the Istituto Italiano di Tecnologia, manipulating a puzzle: in one experiment the robot first takes the puzzle apart (unsolve01) and then solves it (solve01). Data were collected by the AMI research line with the immersive avatar teleoperation stack developed for the ANA Avatar XPRIZE competition, described in the iCub3 avatar system paper (Science Robotics, 2023). Each folder holds Parquet tables of synchronized robot signals, including desired and measured joint states from the walking and manipulation controllers and timestamped RealSense RGB and depth image paths. 22 Parquet files, about 7.7 GB (measured 2026-11-21); read with pandas, pyarrow or Hugging Face datasets. The dataset was made to develop autonomous manipulation algorithms for ergoCub, such as imitation learning of bimanual humanoid manipulation from teleoperated demonstrations. It contains no personal data. Released under CC BY 4.0 by IIT; cite the iCub3 avatar system paper, Science Robotics, doi:10.1126/scirobotics.adh3834.","displayTitle":"ergoCub Humanoid Puzzle Manipulation: Teleoperation Logs of Unsolving and Solving a Puzzle (IIT AMI)","huggingFaceUrl":"https://huggingface.co/datasets/ami-iit/puzzle_manipulation_datasets","id":"9ec938ce-a0ff-5b20-91c1-5e4d5292a8c3","license":"CC-BY-4.0","licenseEvidenceUrl":"https://huggingface.co/datasets/ami-iit/puzzle_manipulation_datasets","objectCountEstimate":24,"probeStatus":"ok","referenceUrl":"https://www.science.org/doi/10.1126/scirobotics.adh3834","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"ami-iit/puzzle_manipulation_datasets\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"ami-iit/puzzle_manipulation_datasets\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=ami-iit, license:cc-by-4.0, size_categories:10K<n<100K, format:parquet, downloads=582"],"format":"mixed","generatedAt":"2026-11-21T22:31:29.695Z","layout":{"example":"solve01/train-00000-of-00012.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"50415231150415c00c15c2094c15c8011500120000a0062873842aa19854d94153a62c05080885c82e050808f9ea30050808bf0c33050808df2e350508080551370508089d7339050808a8953b05080871b73d0508089bd93f0508085afc410508088f1e44050808ac4146050808216248050808be844a050808fea74c0508087ec84e05080809eb500508082d0d53050808e22e55050808be5257050808207359050808a6955b0508086eb75d0508087fdb5f050808befb61050808ec1d64050808ed3f6605080810626805080834846a050808aca66c05080885c86e0508089bea70050808c10c73050808de2e75050808ff50770508082a737905080841957b05080875b77d0508088ed97f050808b1fb81050808d61d84050808fa3f8605080810628805080839848a05080852a68c0508087dc88e05080898ea90050808c00c93050808e72e95050808c952970508087773990508084e959b05080871b79d05080888d99f05080819fca1050808221ea4050808f43fa60508081562a80508088c84aa05080858a6ac050808d6c8ae050808f4eab0050808c20cb3050808e42eb50508088d51b70508082273b90508084795bb0508086ab7bd05080898dbbf050808b2fbc1050808961fc4050808f03fc60508081562c80508083884ca05080857a6cc050808cbc8ce0508089cead0050808bf0cd3050808f22fd5050808","encoding":"hex","key":"solve01/train-00000-of-00012.parquet","truncatedAt":512}],"sampleObjects":["solve01/train-00000-of-00012.parquet","solve01/train-00001-of-00012.parquet","solve01/train-00002-of-00012.parquet","solve01/train-00003-of-00012.parquet","solve01/train-00004-of-00012.parquet"],"sidecars":[{"bytes":"---\ndataset_info:\n- config_name: solve01\n  features:\n  - name: camera/realsense/depth/timestamp\n    dtype: float64\n  - name: timestamps\n    dtype: float64\n  - name: camera/realsense/depth/depth_path\n    dtype: image\n  - name: camera/realsense/rgb/timestamp\n    dtype: float64\n  - name: camera/realsense/rgb/images_path\n    dtype: image\n  - name: walking/com/position/CoM_ZMP_controller/element_0\n    dtype: float64\n  - name: walking/com/position/CoM_ZMP_controller/element_1\n    dtype: float64\n  - name: walking/com/position/CoM_ZMP_controller/element_2\n    dtype: float64\n  - name: walking/com/position/desired/element_0\n    dtype: float64\n  - name: walking/com/position/desired/element_1\n    dtype: float64\n  - name: walking/com/position/desired/element_2\n    dtype: float64\n  - name: walking/com/position/measured/element_0\n    dtype: float64\n  - name: walking/com/position/measured/element_1\n    dtype: float64\n  - name: walking/com/position/measured/element_2\n    dtype: float64\n  - name: walking/com/velocity/desired/element_0\n    dtype: float64\n  - name: walking/com/velocity/desired/element_1\n    dtype: float64\n  - name: walking/com/velocity/desired/element_2\n    dtype: float64\n  - name: walking/dcm/position/desired/element_0\n    dtype: float64\n  - name: walking/dcm/position/desired/element_1\n    dtype: float64\n  - name: walking/dcm/position/measured/element_0\n    dtype: float64\n  - name: walking/dcm/position/measured/element_1\n    dtype: float64\n  - name: walking/dcm/velocity/desired/element_0\n    dtype: float64\n  - name: walking/dcm/velocity/desired/element_1\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_0\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_1\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_2\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_3\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_4\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_5\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_6\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_7\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_8\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_9\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_10\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_11\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_12\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_13\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_14\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_15\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_16\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_17\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_18\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_19\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_20\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_21\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_22\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_23\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_24\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_25\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_26\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_27\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_28\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_29\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_30\n    dtype: float64\n  - name: walking/joints_state/positions/desired/element_31","encoding":"utf8","key":"README.md","name":"README.md","size":119060,"truncated":true}],"sourceObjectKey":"solve01/train-00000-of-00012.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"ami-iit/puzzle_manipulation_datasets\")\nprint(ds)","starterSamplePath":"solve01/train-00000-of-00012.parquet","totalObjects":24},"summary":"Logged teleoperation data from ergoCub, the humanoid robot of the Istituto Italiano di Tecnologia, manipulating a puzzle: in one experiment the robot first takes the puzzle apart…","title":"ergoCub Humanoid Puzzle Manipulation: Teleoperation Logs of Unsolving and Solving a Puzzle (IIT AMI)","updatedAt":"2026-11-22T00:55:25.710Z","visibility":"public","websiteUrl":"https://ergocub.eu/"}}