{"version":1,"type":"story","url":"https://digestai.news/story/daimon-robotics-builds-tactile-data-pipeline-for-robot-manipulation","json":"https://digestai.news/story/daimon-robotics-builds-tactile-data-pipeline-for-robot-manipulation.json","markdown":"https://digestai.news/story/daimon-robotics-builds-tactile-data-pipeline-for-robot-manipulation.md","slug":"daimon-robotics-builds-tactile-data-pipeline-for-robot-manipulation","headline":"Daimon Robotics builds tactile data pipeline for robot manipulation","summary":"Daimon Robotics is creating an end‑to‑end pipeline that links tactile sensors, large‑scale data collection, simulation and AI models to give robots a sense of touch. Partner & VP of Technology Dr. Zhu says the company is positioning itself as an infrastructure provider rather than a foundation‑model developer. The system records tactile feedback at 30 Hz and aims to gather more than 10,000 hours of contact data, with a longer‑term Daimon‑Infinity goal of over 1 million hours.\n\nThe firm is calibrating real‑world measurements against simulated environments to narrow the sim‑to‑real gap, which Zhu described as “kind of terrible” today. Daimon’s Umi device captures human‑hand interactions, and the collected data feeds a tactile‑grounded world model called Daimon‑TWM. Demonstrations at IROS 2026 showed the technology handling delicate assembly, packaging broken glass and other manipulation tasks. Lenovo Capital and Inovance Technology are identified as key partners, and the company reports moving beyond proof‑of‑concept projects into industrial assembly lines, though specific customers remain confidential.","keyPoints":["Daimon aims to collect over 10,000 hours of tactile data, targeting 1 million hours under the Daimon‑Infinity initiative.","The pipeline captures tactile feedback at 30 Hz and links real‑world measurements to simulation for robot training.","Partners include Lenovo Capital and Inovance Technology; demonstrations at IROS 2026 showed assembly and glass‑packaging tasks."],"whyItMatters":"Providing scalable tactile data and simulation bridges a key gap in robot manipulation, enabling more reliable contact‑rich tasks across industry.","category":{"slug":"robotics","name":"Robotics & Physical AI","url":"https://digestai.news/category/robotics"},"entities":{"companies":["Daimon Robotics","Lenovo Capital","Inovance Technology"],"models":["Daimon-TWM"],"people":["Dr. Zhu"]},"firstPublishedAt":"2026-10-19T17:07:52Z","updatedAt":"2026-10-19T17:07:52Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"The Robot Report","title":"Daimon Robotics building the infrastructure layer for tactile intelligence","url":"https://therobotreport.com/daimon-robotics-building-the-infrastructure-layer-for-tactile-intelligence","publishedAt":"2026-10-19T17:07:52Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Daimon Robotics builds tactile data pipeline for robot manipulation\", 19 October 2026, https://digestai.news/story/daimon-robotics-builds-tactile-data-pipeline-for-robot-manipulation","publisher":"Digest AI","title":"Daimon Robotics builds tactile data pipeline for robot manipulation","datePublished":"2026-10-19T17:07:52Z","url":"https://digestai.news/story/daimon-robotics-builds-tactile-data-pipeline-for-robot-manipulation"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}