计算机科学
正确性
运行时验证
概率逻辑
分布式计算
自动机
过程(计算)
模型检查
形式验证
理论计算机科学
算法
程序设计语言
人工智能
作者
Mu Yongan,Wei Liu,Tao Lü,Juan Li,Shen Gao,Zihao Wang
出处
期刊:Journal of Intelligent and Fuzzy Systems
[IOS Press]
日期:2023-12-02
卷期号:45 (6): 10305-10322
摘要
The self-adaptive multi-agent system requires adaptive adjustments based on the dynamic environment during its runtime. Heterogeneous agent can accomplish different task goals, enhance the efficiency of system operation, but its complex collaboration problem poses new challenges to the study of verification of adaptive policies for heterogeneous multi-agents. This paper proposes a runtime verification method for self-adaptive multi-agent systems using probabilistic timed automata. The method constructs a probabilistic timed automaton model by formally describing the functional characteristics of heterogeneous agents and integrating random factors in the environment to simulate the operation process of the self-adaptive multi-agent system. Regarding the collaboration logic among heterogeneous agents, security constraints are established to ensure the security of state transition processes during system operation. Combining model checking with runtime quantitative verification methods to conduct experiment and applying it in the case of an intelligent unmanned parking system. Experimental results manifest the correctness of the cooperation logic between agents can effectively ensure the stability of the system at runtime. Significant improvement in system uptime and efficiency compared to the initial system without runtime quantitative validation.
科研通智能强力驱动
Strongly Powered by AbleSci AI