首页|Findings from Shandong University Broaden Understanding of Robotics (Flingflow: Llm-driven Dynamic Strategies for Efficient Cloth Flattening)
Findings from Shandong University Broaden Understanding of Robotics (Flingflow: Llm-driven Dynamic Strategies for Efficient Cloth Flattening)
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
Investigators publish new report on Ro botics. According to news reporting originating from Jinan, People's Republic of China, by NewsRx correspondents, research stated, "The proficiency of robots in cloth manipulation is crucial for their potential widespread deployment in hous ehold service contexts, with the task of unfolding cloth being particularly indi spensable. Unlike rigid objects, cloth has a high-dimensional state space, which poses significant challenges for robotic operations." Funders for this research include High speed and high precision industrial robot , National Natural Science Foundation of China (NSFC), Meituan Academy of Roboti cs Shenzhen. Our news editors obtained a quote from the research from Shandong University, "T his paper presents a robotic framework that integrates dynamic and static operat ions for cloth unfolding. Dynamic operations are introduced in a single-arm scen ario, employing gravity to expedite flattening. Initially, we define the classif ication of cloth states and operational skills. Subsequently, in skill selection , a Large Language Model (LLM) is utilized to make decisions based on the curren t state, selecting skills appropriate for the given situation. For the determina tion of operation points, a cloth region segmentation network extracts key featu res of the cloth, and the final operation points are determined through geometri c analysis of the masks. Experiments on a real robot demonstrate that our method can successfully unfold cloths of various initial conditions, colors, sizes, te xtures, shapes and materials, achieving over 95$\ %$ coverage-defined as the ratio of the current are a of the fabric to its fully expanded area-thereby proving the effectiveness o f the combined dynamic and static operation strategy."
JinanPeople's Republic of ChinaAsiaEmerging TechnologiesMachine LearningRoboticsRobotsShandong University
NETL
NSTL
