首页|Recent Findings in Robotics Described by Researchers from University of Minnesot a (Talk Through It: End User Directed Manipulation Learning)
Recent Findings in Robotics Described by Researchers from University of Minnesot a (Talk Through It: End User Directed Manipulation Learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Fresh data on Robotics are presented i n a new report. According to news originating from Minneapolis, Minnesota, by Ne wsRx correspondents, research stated, "Training robots to perform a huge range o f tasks in many different environments is immensely difficult. Instead, we propo se selectively training robots based on end-user preferences." Financial support for this research came from Minnesota Robotics Institute. Our news journalists obtained a quote from the research from the University of M innesota, "Given a factory model that lets an end user instruct a robot to perfo rm lower-level actions (e.g. ‘Move left'), we show that end users can collect de monstrations using language to train their home model for higher-level tasks spe cific to their needs (e.g. ‘Open the top drawer and put the block inside'). We d emonstrate this framework on robot manipulation tasks using RLBench environments . Our method results in a 13 % improvement in task success rates co mpared to a baseline method. We also explore the use of the large vision-languag e model (VLM), Bard, to automatically break down tasks into sequences of lower-l evel instructions, aiming to bypass end-user involvement."
MinneapolisMinnesotaUnited StatesNorth and Central AmericaEmerging TechnologiesMachine LearningNano-robotR obotRoboticsUniversity of Minnesota
NETL
NSTL
