论文部分内容阅读
针对室内移动机器人在智能服务任务中难以获得复杂环境语义的问题,通过设计云端语义库,实现基于语义获取框架的机器人语义地图构建,使机器人不仅掌握面向导航的环境几何描述,而且获得了复杂环境下基于丰富语义库的含物品关联归属关系的语义地图,解决了语义地图构建过程中语义信息添加可靠性低、地图更新存在误差及扩展性不足等问题.首先给出了一种语义库构建方案,基于支持向量机实现语义库分类形成子语义库,在子语义库基础上基于网络文本分类来提取关键特征点形成特征模型库,通过语义分类列表整合子语义库实现物品查询.其次,论述了面向智能服务任务的云端语义地图实现,基于多尺度图像分割与视差图分析,设计标注库与归属库描述物品关联归属关系.最后进行了有关语义地图构建及语义库分类效率的仿真实验与结果分析,验证了方法的有效性.
In order to solve the problem that indoor mobile robots can not obtain complex environment semantics easily in intelligent service tasks, a semantic access framework based robotic semantic map is constructed by designing a cloud semantic library so that the robot not only grasps the navigational environment geometric description but also obtains the complex environment A semantic map based on the rich semantic library is proposed to solve the problem of low reliability of semantic information in the process of constructing the semantic map, errors in the update of the map and lack of extensibility.Firstly, a semantic library construction scheme , Based on support vector machine to realize the semantic library classification to form the sub-semantic library, based on the sub-semantic library, based on the network text classification to extract the key feature points to form the feature model library, and through the semantic classification list integration sub-semantic library to achieve article query.Secondly, Based on the multi-scale image segmentation and parallax map analysis, the relationship between the label library and the attribution library is described. Finally, the semantic map construction and the semantic library classification efficiency are simulated and the results are analyzed , Verified the method Effectiveness.