Formal Modeling of Intelligent Model with Enhanced State Transitions and Hybrid Module Semantics
B. X. Yu, Tong Yang, Jiayu Tong, Xi Liu, Mingsheng Wang · 2025
Formal modeling is the most important step in Model-Based Systems Engineering (MBSE). The existing engineering mainly focuses on modeling and simulation of non-intelligent entities. With the development of artificial intelligence technology, more and more intelligent models are involved in the industry, making it difficult for traditional modeling methods to meet the needs of formal modeling of intelligent models. In view of the above shortcomings, this paper analyzes the common core intelligence characteristics of current intelligent models, and proposes a formal modeling extension method for intelligent models. Two new types of modules, TypeBlock and IntelligenceBlock, are designed to enhance the description ability of traditional functional modules and intelligent units. Aiming at the state transition of the state machine, some features such as interactive combination state, increasing priority mechanism and dynamic threshold adjustment are designed to enhance the modeling ability of complex state transition involving intelligent models. Finally, this paper uses the unified functional model unit transformation framework to transform the intelligent model into FMU, and realizes efficient modeling and simulation through the integration of intelligent model and multi-domain simulation platform. Through case verification, our method effectively solves the shortcomings of existing methods in the formal modeling of intelligent systems, provides a new idea for the design and verification of intelligent systems, and promotes the application and development of MBSE in intelligent systems engineering.