论文部分内容阅读
为了实现机器人理解自然语言的目的,提出了一种应用于中文服务指令的自主处理方法,旨在将中文服务指令直接映射为动作序列.首先深入地分析收集的中文服务指令,提炼出了关键信息与句法结构的对应关系.在此基础上,以短语组块的语法规则为基础,利用概率模型提取指令中的关键信息.为了解决指令解析问题,设计了一种服务任务分解模板,机器人依据该模板可以将中文服务指令直接映射为动作序列.最后进行仿真实验,统计了中文服务指令自主处理过程中各个环节的实验结果,其中组块标注、信息提取及指令解析的准确率平均值分别达到了92.9%、92.7%和97.2%,验证了本文方法的有效性和可行性.
In order to realize the purpose of robot to understand natural language, an autonomous processing method applied to Chinese service instructions is proposed, which aims to directly map Chinese service instructions into action sequences.Firstly, the collected Chinese service instructions are analyzed in depth and the key information And the syntactic structure.On this basis, based on the grammar rules of the phrase chunk, the key information in the instruction is extracted by using the probability model.In order to solve the problem of instruction parsing, a service task disassembly template is designed, The template can directly map the Chinese service instructions into action sequences.Finally, the simulation experiments are carried out, and the experimental results of each link in the process of Chinese service instruction autonomy are calculated, in which the mean values of chunk annotation, information extraction and instruction resolution respectively reach 92.9%, 92.7% and 97.2%, respectively, which verify the effectiveness and feasibility of the proposed method.