{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "mandlekar-zhu-garg-et-al-roboturk-crowdsourced-teleoperation-2018", "title": "RoboTurk: collecting simulated robot demonstrations with a phone", "summary": "A person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO.", "boundary": "The 2018 contribution uses simulated robots. A later real-robot extension has different authors and results and is not included in this award. The interface comparison involves eight students; failing to detect a difference from a VR controller does not prove equivalence. Cross-Pacific trials take longer. The paper gives inconsistent 20- versus 22-hour platform totals, so no reconciled exact total is asserted here. Demonstrations come from human workers; aggregate trajectory hours are not elapsed time. Learning returns are not success percentages, and benefits are not monotonic for every demonstration count.", "area": "Robot learning and remote demonstration infrastructure", "topic": "Artificial intelligence", "year": 2018, "lane": "HUMAN", "result": "ALGORITHM", "package_id": "mandlekar-zhu-garg-et-al-roboturk-crowdsourced-teleoperation-2018", "target": { "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "content_hash": "sha256:502a01958020bf371363ce061d026d434820e0f67ab554370f16ddfd043f7104" }, "humans": [ "Ajay Mandlekar", "Yuke Zhu", "Animesh Garg", "Jonathan Booher", "Max Spero", "Albert Tung", "Julian Gao", "John Emmons", "Anchit Gupta", "Emre Orbay", "Silvio Savarese", "Li Fei-Fei" ], "authors_complete": true, "human_role_note": "Source-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "cur20261017-arxiv-1811-02790v1", "title": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1", "date": "2018-11-07" } ], "points": "100000", "published_grade": { "id": "mandlekar-zhu-garg-et-al-roboturk-crowdsourced-teleoperation-2018", "entry_id": "mandlekar-zhu-garg-et-al-roboturk-crowdsourced-teleoperation-2018", "band": "A", "modifier": null, "target": { "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "content_hash": "sha256:502a01958020bf371363ce061d026d434820e0f67ab554370f16ddfd043f7104" }, "what_changed": "A person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO.", "scope": "A person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO. The 2018 contribution uses simulated robots. A later real-robot extension has different authors and results and is not included in this award. The interface comparison involves eight students; failing to detect a difference from a VR controller does not prove equivalence. Cross-Pacific trials take longer. The paper gives inconsistent 20- versus 22-hour platform totals, so no reconciled exact total is asserted here. Demonstrations come from human workers; aggregate trajectory hours are not elapsed time. Learning returns are not success percentages, and benefits are not monotonic for every demonstration count.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "content_hash": "sha256:502a01958020bf371363ce061d026d434820e0f67ab554370f16ddfd043f7104" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-23T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Ajay Mandlekar", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Yuke Zhu", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Animesh Garg", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jonathan Booher", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Max Spero", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Albert Tung", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Julian Gao", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "John Emmons", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Anchit Gupta", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Emre Orbay", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Silvio Savarese", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Li Fei-Fei", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "The 2018 contribution uses simulated robots. A later real-robot extension has different authors and results and is not included in this award. The interface comparison involves eight students; failing to detect a difference from a VR controller does not prove equivalence. Cross-Pacific trials take longer. The paper gives inconsistent 20- versus 22-hour platform totals, so no reconciled exact total is asserted here. Demonstrations come from human workers; aggregate trajectory hours are not elapsed time. Learning returns are not success percentages, and benefits are not monotonic for every demonstration count.", "domain": "Robot learning and remote demonstration infrastructure" }, "sources": [ { "citation": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1" } ], "statement": "# RoboTurk: collecting simulated robot demonstrations with a phone\n\nA person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO.\n\n## Scientific scope\n\nThe 2018 contribution uses simulated robots. A later real-robot extension has different authors and results and is not included in this award. The interface comparison involves eight students; failing to detect a difference from a VR controller does not prove equivalence. Cross-Pacific trials take longer. The paper gives inconsistent 20- versus 22-hour platform totals, so no reconciled exact total is asserted here. Demonstrations come from human workers; aggregate trajectory hours are not elapsed time. Learning returns are not success percentages, and benefits are not monotonic for every demonstration count.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nAjay Mandlekar; Yuke Zhu; Animesh Garg; Jonathan Booher; Max Spero; Albert Tung; Julian Gao; John Emmons; Anchit Gupta; Emre Orbay; Silvio Savarese; Li Fei-Fei\n\nSource-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred.\n\n\n\n## Original sources\n\n- [RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation](https://arxiv.org/abs/1811.02790v1)\n", "summary": "A person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO.", "title": "RoboTurk: collecting simulated robot demonstrations with a phone" }, "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "source_ref": { "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "content_hash": "sha256:502a01958020bf371363ce061d026d434820e0f67ab554370f16ddfd043f7104" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:00b7e66a5e178e0def771ea4a2f19b5d0bac1d24e0c43692df60f85688536544" }, "target": { "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "content_hash": "sha256:502a01958020bf371363ce061d026d434820e0f67ab554370f16ddfd043f7104" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "mandlekar-zhu-garg-et-al-roboturk-crowdsourced-teleoperation-2018", "entry_id": "mandlekar-zhu-garg-et-al-roboturk-crowdsourced-teleoperation-2018", "revision": 1, "title": "RoboTurk: collecting simulated robot demonstrations with a phone", "context": "Robot learning and remote demonstration infrastructure", "track": "Reusable scientific method, experimental result or historical framework", "band": "A", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-23", "evidence_cutoff": "2026-10-23", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "A person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO.", "scope": "A person can demonstrate a robot-arm movement by moving a smartphone while watching a video stream. RoboTurk connects that familiar input to cloud-hosted robot simulations through low-latency communication. The 2018 platform collects a pilot set of more than 2,200 successful demonstrations and studies whether those examples help reinforcement learning explore sparse-reward tasks. Its contribution is the teleoperation and collection pipeline with its documented learning use, separate from the earlier ideas of imitation learning, WebRTC and PPO. The 2018 contribution uses simulated robots. A later real-robot extension has different authors and results and is not included in this award. The interface comparison involves eight students; failing to detect a difference from a VR controller does not prove equivalence. Cross-Pacific trials take longer. The paper gives inconsistent 20- versus 22-hour platform totals, so no reconciled exact total is asserted here. Demonstrations come from human workers; aggregate trajectory hours are not elapsed time. Learning returns are not success percentages, and benefits are not monotonic for every demonstration count.", "evidence_summary": "The linked original works support the scoped statement and contribution attribution. WRS has not independently replayed experiments or reconstructed a full mathematical proof. The 2018 contribution uses simulated robots. A later real-robot extension has different authors and results and is not included in this award. The interface comparison involves eight students; failing to detect a difference from a VR controller does not prove equivalence. Cross-Pacific trials take longer. The paper gives inconsistent 20- versus 22-hour platform totals, so no reconciled exact total is asserted here. Demonstrations come from human workers; aggregate trajectory hours are not elapsed time. Learning returns are not success percentages, and benefits are not monotonic for every demonstration count.", "references": [ { "title": "RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation", "url": "https://arxiv.org/abs/1811.02790v1", "locator": "Original scoped contribution, statement/method and evidence limits; source-specific attribution retained." } ], "roles": [ { "who": "The named authors in the research record", "credit": "SCIENCE", "did": "Source-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred." } ], "target": { "record_id": "urn:uuid:b0571efe-dedc-4655-add8-1686486b2514", "version": 1, "content_hash": "sha256:502a01958020bf371363ce061d026d434820e0f67ab554370f16ddfd043f7104" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "100000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }