| 摘要: |
| 针对目标识别专家系统在工程应用中可维护性和可移植性差的问题，研究了一种基于目标识别知识模型的推理过程可柔性重组的专家系统。以识别要素项表征识别知识的元数据，利用识别要素项配置源数据格式和识别模板，当部署位置、侦察手段以及侦察数据格式改变时，只需要修改源数据配置、识别模板配置和识别知识库，而不需要修改软件代码，就能满足不同任务的识别需求。利用3个典型场景的应用测试验证了方法的可行性。 |
| 关键词: 目标识别 专家系统 知识模型 柔性重组 |
| DOI： |
|
| 基金项目:国家重点研发计划项目（2017YFC1404900） |
|
| An expert system for target recognition based on knowledge model |
| HE Wenjiao |
| (Southwest China Institute of Electronic Technology,Chengdu 610036,China) |
| Abstract: |
| In view of the poor maintainability and poor portability of the target recognition expert system in engineering application,an expert system which can be reorganized flexibly based on the knowledge model is studied.By using the identification metadata configuration source data and identification template,it supports the dynamic reorganization of the reasoning process.When the application scenario,deployment location,reconnaissance methods,and even the reconnaissance data format change,only the configuration file needs to be modified,not the system code.The application tests of three typical scenarios verify the feasibility of the method. |
| Key words: target recognition expert system knowledge model flexible reorganization |
