• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (08): 1510-1516.

• 论文 • 上一篇    下一篇

无人驾驶汽车决策系统的规则描述与代码生成方法

兰韵1 ,刘万伟1,董威1,刘斌斌1,付辰2,刘大学3   

  1. 3(1.国防科学技术大学计算机学院,湖南 长沙 410073;2.中国科学院软件所计算机科学国家重点实验室,北京 100000;3.国防科学技术大学机电工程与自动化学院,湖南 长沙 410073)
  • 收稿日期:2014-10-28 修回日期:2014-12-16 出版日期:2015-08-25 发布日期:2015-08-25
  • 基金资助:

    国家973计划资助项目(2009CB723803);国家自然科学基金资助项目(60873120,61103012)

Research on rule editing and code generation for
the high-level decision system of unmanned vehicles  

LAN Yun1,LIU Wanwei1,DONG Wei1,LIU Binbin1,FU Chen2,LIU Daxue3   

  1. (1.College of Computer,National University of Defense Technology,Changsha 410073;2.State Key Laboratory of Computer Science,Institute of Software,Chinese Academy of Sciences,Beijing 100000;3.College of Electric Engineering and Automotive,National University of Defense Technology,Changsha 410073,China)
  • Received:2014-10-28 Revised:2014-12-16 Online:2015-08-25 Published:2015-08-25

摘要:

无人驾驶汽车的高层决策系统是其实现自主驾驶功能的核心,在决策系统的开发过程中,由于信号变量与状态数量庞大等原因,系统面临着可维护性与可拓展性差的难题。UNMANNED_RULE_EDIT是面向无人车高层决策系统的规则编辑与代码自动生成辅助工具,具有图形化编辑和代码自动生成的功能,有利于开发人员清晰直观地设计规则,并摆脱繁琐的编码工作。主要讨论了UNMANNED_RULE_EDIT的中间语言(规则语言)设计和代码自动生成算法。目前,该工具已初步应用在国内某单位研制的无人车上,为其决策系统的开发提供便利。

关键词: 无人驾驶汽车, 决策系统, 规则语言, 代码自动生成

Abstract:

High-level decision systems are the central module of unmanned vehicles. Due to the extremely large amount of states and signals, the development process of the intelligence decision system is very complex, and maintaining such a system will encounter great difficulties. To overcome these problems, we design and implement a rule editing and code generation tool, called UNMANNED_RULE_EDIT.  It features GUIbased editing and automatic code generation and can help developers to generate controlling codes via editing a series of statetransition rules rather than writing codes, thus great succinctness and intuitiveness can be yielded. The paper mainly focuses on the discussion of the rule language and the code generation algorithm. As a preliminary application, this tool has been used in an unmanned vehicle system in China and facilitated the development of its decision system.

Key words: unmanned vehicles;intelligence decision system;rule language;code generation