FurnMove

该数据集名为FurnMove，是一项创新的多智能体任务，智能体需协同工作，将一件家具从客厅移动至指定目标位置，这一过程要求智能体在每一个时间步上都进行协调。该数据集包含30个客厅场景，其中20个用于训练，5个用于验证，另外5个用于测试。重点在于将电视作为搬运对象，并将电视柜作为目标位置。规模上，数据集跨越5个场景，共1000个剧集。任务类型为多智能体协作移动家具。

This dataset, named FurnMove, is an innovative multi-agent task that requires agents to collaborate to move a piece of furniture from a living room to a designated target position, with mandatory coordination between agents at every time step. The dataset includes 30 living room scenarios, 20 of which are allocated for training, 5 for validation, and the remaining 5 for testing. The core focus of the task is to take a television as the moved item, with a TV stand as the target location. In terms of scale, the dataset contains 1000 episodes across 5 scenarios, and the task type is multi-agent collaborative furniture movement.