A Dataset Schema for Cooperative Learning from Demonstration in a Multi-robots System
收藏arXiv2019-12-04 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/1912.01741v1
下载链接
链接失效反馈官方服务：
资源简介：
本数据集旨在支持多机器人系统中的协作学习示范，由巴伊亚州立大学(UNEB)的研究人员开发。数据集专注于从领域专家的示范中学习协调行为，特别是在多机器人系统中。数据集内容包括从示范中提取的多种非标量数据类型，如布尔条件、对象列表和步骤列表，用于描述机器人的协作计划。创建过程涉及使用SPlanner工具生成机器人足球比赛中的协作计划，并通过模糊C均值(FCM)算法进行组织。该数据集的应用领域是机器人足球，旨在通过学习领域专家的示范来提高机器人在比赛中的协作和决策能力。
This dataset, developed by researchers from the State University of Bahia (UNEB), aims to support collaborative learning demonstrations in multi-robot systems. It focuses on learning coordinated behaviors from demonstrations provided by domain experts, particularly within multi-robot systems. The dataset contains various non-scalar data types extracted from these demonstrations, such as boolean conditions, object lists and step lists, which are used to describe the collaborative plans of robots. The dataset creation process involved using the SPlanner tool to generate collaborative plans for robot soccer matches, and organizing these plans via the Fuzzy C-Means (FCM) algorithm. The application field of this dataset is robot soccer, with the goal of improving the collaborative and decision-making capabilities of robots during matches by learning from domain expert demonstrations.
提供机构：
巴伊亚州立大学(UNEB)
创建时间：
2019-12-04
