摘要
无人战争将会成为未来主流的战争形式,跨平台异构协同也将成为作战指挥系统控制的核心理念。现有的指控平台通常为传统的键盘输入,对指令员的专业要求高,且无法实现对跨平台异构设备的协同控制。基于这两点问题,运用日趋成熟的语音识别技术,为无人作战控制系统设计了基于语音控制的人机交互模式,为实现跨平台异构设备的协同控制,采用平台无关的标准接口描述语言(IDL)作为系统的输出指令。对于语音控制平台的设计,首先确定了可用于跨平台异构协同的目标指令的输出形式为标准化IDL,并基于实验范围与IDL标准确定了语音识别边界;其次采用科大讯飞语音识别技术作为语音命令向文本指令的转化工具;然后将文本化的指令利用ddpaser进行断句分析;最后,经过语言支撑集数据库、标准化指令数据库、IDL指令数据库进行标准化匹配,获得边界范围内的标准化IDL作为指令输出。研究将为作战指控系统的开发设计,实现人机语音交互与跨平台异构控制提供思路。
Unmanned war will become the mainstream form of war in the future,and cross-platform heterogeneous cooperation will also become the core concept of battle command system control.The traditional keyboard input is usually used in the existing accusation platform,which requires high professional qualities of the command operator,cannot realize the cooperative control of cross-platform heterogeneous devices.Based on these two problems,the speech recognition technology,which becoming mature in recent years,is used in this paper to design a human-computer interaction mode based on voice control for the unmanned combat control system.In order to achieve the collaborative control of cross-platform heterogeneous devices,Interface Description Language(IDL),a platform-independent standard,is used as the output instruction of the system.For the design of the speech control platform,the output form of the target instruction which can be used for cross-platform heterogeneous collaboration is determined as standardized IDL,and the boundary of speech recognition is determined based on the experimental range and IDL standard.Secondly,the speech recognition technology of iFlytek is used as the conversion tool from speech command to text command.Then,the textual instruction is analyzed by using DDPASER.Finally,standardized matching is carried out by language support set database,standardized instruction database and IDL instruction database to obtain the standardized IDL within the boundary range as instruction output.The ideas for the development and design of combat command system and the realization of speech human-computer interaction and cross-platform heterogeneous control are provided.
作者
王开宇
张弘瑞
李聪
王献昌
WANG Kaiyu;ZHANG Hongrui;LI Cong;WANG Xianchang(College of Computer Science and Technology,Jilin University,Changchun 130012,China;College of Software,Jilin University,Changchun 130012,China;Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education,Jilin University,Changchun 130012,China;Chengdu Kestrel Artificial Intelligence Research Institute,Chengdu 611730,China)
出处
《无人系统技术》
2021年第4期77-84,共8页
Unmanned Systems Technology
基金
国防科技创新特区“跨异构平台互联互通互操作关健理论与支撑项目”。
关键词
跨平台异构
集群协同
自然语言识别
断句分析
标准接口描述语言
模糊匹配
Cross-platform Heterogeneity
Cluster Collaboration
Natural Language Recognition
Sentence Breaking Analysis
Standard Interface Description Language
Fuzzy Matching
