Integrating AI Microservices into Hard-Real-Time SoS to Ensure Trustworthiness of Digital Enterprise Using Mission Engineering

Abstract

Due to the increased complexities and operating speeds of today’s and tomorrow’s system-of-systems (SoS) architectural configurations for digital enterprises, the design of new domain architecture management systems is required. A key element of these new designs will be the incorporation of Artificial Intelligence (AI) microservices to provide dynamically containerized and orchestrated service capabilities within a lightweight interoperability fabric with the ability to operate in hard-real-time environments. Each containerized AI microservice exposes an independent, programmable function, which enables it to be easily reused, evolved, or replaced without compromising interoperability across critical mission essential functions to execute mission threads. In addition, embedded in this design needs to be a trust management layer to enforce reliable messaging and trust amongst the actors. This paper provides a framework for planned research and demonstrates the feasibility of microservices using a representative simple problem to demonstrate the application of the framework. Early positive analysis results using AI microservices within an SoS environment shows that the 500 milli-second (ms) threshold for latency can be met.