摘要
针对现有虚拟装配系统普遍缺乏对复杂情境的分析能力,难以根据上下文有效预测用户动作意图的问题,给出了虚拟装配中情境上下文的形式化定义,建立了智能虚拟装配动作预测模型。该模型通过对装配动作模式的挖掘,进行基于实例的动作意图预测,实现系统的智能虚拟装配。结合水轮发电机组的转轮体装配实例,给出了模型的实现框架。通过实验验证,该模型能够根据情境上下文准确预测用户的动作意图,实时调整零部件的动作以自动完成装配。
Traditional virtual assembly system lacks the ability of analyzing the complicated situation and predicting user＇s action intention with the context effectively. In order to solve problems above, the formal definition of context is gaven and an intelligent action prediction model is proposed in the virtual assembly. Through mining the assembly action mode the action intention based on the examples is predicted, and intelligent virtual assembly is realized. Furthermore, the framework of the model is put forward by taking assembly of the runner hub in the hydroelectric generating set as a example. Through the experiment, the model can accurately predict the user＇s action intention with the context, adjust the action of the parts in real time to finish the assembly automatically.
出处
《计算机工程与设计》
CSCD
北大核心
2013年第4期1366-1370,共5页
Computer Engineering and Design
关键词
虚拟装配
情境上下文
动作模式
意图预测
模型
框架
virtual assembly
context awareness
action mode
intention prediction
model
framework
