系统仿真学报 ›› 2026, Vol. 38 ›› Issue (4): 829-854.doi: 10.16182/j.issn1004731x.joss.25-0797

• 专家约稿 •    下一篇

基于模型的系统验证:理论框架、关键技术及未来展望

孙波1, 任羿2, 王偲霖3, 刘奇4, 李志栋1   

  1. 1.北京空间飞行器总体设计部,北京 100094
    2.北京航空航天大学,北京 100191
    3.哈尔滨工程大学,黑龙江 哈尔滨 150001
    4.苏州同元软控信息技术有限公司,江苏 苏州 215000
  • 收稿日期:2025-08-22 修回日期:2025-11-07 出版日期:2026-04-20 发布日期:2026-04-22
  • 通讯作者: 任羿
  • 第一作者简介:孙波(1974-),男,研究员,博士,研究方向为数字试验与测试,装备健康管理,智能装备可信性测试与评估。
    孙波 研究员(博士生导师),毕业于北京航空航天大学,获计算机软件博士学位,现任中国航天科技集团有限公司第五研究院总体部研究员,中国仿真学会理事,中国仿真学会数字试验与测试专业委员会主任委员,中国造船工程学会舰船保障学术委员会委员、指挥控制协会集群智能与协同控制专业委员会委员。从事基于模型的测试验证、数字化试验与测试、健康管理等方面的研究工作。出版专著一部。
    任羿 北京航空航天大学蓝天杰出教授(二级),可靠性系统工程学科方向带头人。入选国家级领军人才计划,兼任中国仿真学会理事、中国仿真学会数字试验与测试分委会副主任、系统仿真学报编委。发表SCI论文100余篇,授权发明专利80余项,主编中英文专著3部,主持多项国家级重大重点科研项目。获国家科技进步二等奖1项、省部级科技进步一等奖4项、集体荣获(带头人)省部级科技创新团队奖1项。

Model-based System Verification: Theoretical Framework, Key Technologies, and Future Prospects

Sun Bo1, Ren Yi2, Wang Silin3, Liu Qi4, Li Zhidong1   

  1. 1.Beijing Institute of Spacecraft System Engineering, Beijing 100094, China
    2.Beihang University, Beijing 100191, China
    3.Harbin Engineering University, Harbin 150001, China
    4.Suzhou Tongyuan Soft Control Information Technology Co. , Ltd. , Suzhou 215000, China
  • Received:2025-08-22 Revised:2025-11-07 Online:2026-04-20 Published:2026-04-22
  • Contact: Ren Yi

摘要:

为解决传统的系统验证方法在效率、覆盖度和可追溯性等方面面临的严峻挑战,提出基于模型的系统验证(model-based system verification,MBSV),将验证活动深度融入基于模型的系统工程模型体系及其演化过程。给出了MBSV的构建逻辑,提出一种以系统建模语言(system modeling language,SysML)为基础,融合需求、结构、行为与约束的多视图统一验证建模策略。论述了旨在提升验证效率的路径代表性选择与等价类约简算法、测试路径搜索原理,以及基于RL的路径智能搜索机制;描述了从抽象路径到可执行测试用例的自动化生成流程,探讨了基于功能模型单元与仿真平台SDK的双通道自动化执行机制,以实现验证任务在异构仿真环境中的无缝部署与闭环反馈。展望了MBSV未来发展的重点方向。

关键词: 基于模型的系统验证, 基于模型的系统工程, 系统建模语言, 测试路径搜索, 测试路径约简, 测试用例自动生成, 测试仿真平台, 姿态确定与控制系统

Abstract:

Traditional system verification methods face significant challenges in terms of efficiency, coverage, and traceability. To address these issues, this paper introduced model-based system verification (MBSV), which deeply integrated verification activities within the model-based systems engineering model system and evolution process. It presented the foundational logic of MBSV and proposed a multi-view unified verification modeling strategy based on system modeling language (SysML), integrating requirements, structure, behavior, and constraints. The paper discussed the algorithms for selecting representative paths and reducing equivalent classes to enhance verification efficiency, the principles of test path search, as well as the intelligent path search mechanism based on RL. It detailed the automated generation process for transforming abstract paths into executable test cases and discussed a dual-channel automated execution mechanism based on functional mock-up units and simulation platform SDKs to enable the seamless deployment and closed-loop feedback of verification tasks within heterogeneous simulation environments. The paper outlined future research directions for MBSV.

Key words: model-based system verification, model-based systems engineering, system modeling language, test path search, test path reduction, automated test case generation, test simulation platform, attitude determination and control system

中图分类号: