首页> 外国专利> METHOD FOR RECOMMENDING ROBOT ALTERNATIVE ACTION USING WEIGHTED ACTION-COUPLED SEMANTIC NETWORK

METHOD FOR RECOMMENDING ROBOT ALTERNATIVE ACTION USING WEIGHTED ACTION-COUPLED SEMANTIC NETWORK

机译:加权动作耦合语义网络的机器人交替动作推荐方法

摘要

PURPOSE: A determination method of an alternative action of a robot using a weighted action-coupled semantic technology is provided to exclude complicated inference procedure, add the weight to the associative relationship between predetermined data, and rapidly select an alternative action. CONSTITUTION: The data structure in which the individual weighted value is granted is formulated. The goal object and goal action are determined from the command for execution. Among the items which are related with the goal object, the confirmed item in an input image is determined as the neighboring item. Among the items which are related with the neighboring item, the item of which the weighted value is the most related is determined as the winner node. Among the action names which are related with the winner node, the action name of which the weighted value is the most related is determined as the alternative action.
机译:目的:提供一种使用加权动作耦合语义技术确定机器人替代动作的方法,以排除复杂的推理过程,将权重添加到预定数据之间的关联关系中,并快速选择替代动作。构成:制定了授予个人加权值的数据结构。从执行命令确定目标对象和目标动作。在与目标对象有关的项目中,将输入图像中的确认项目确定为相邻项目。在与相邻项目相关的项目中,加权值最相关的项目被确定为获胜者节点。在与获胜者节点相关的动作名称中,将加权值最相关的动作名称确定为替代动作。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号