What is it about?
Training modern AI requires lots of high-quality data, yet obtaining this data in the actual world can be expensive and difficult. One way is to employ AI simulations, in which simulated environments enable AI to learn and improve safely. Digital twins—detailed virtual replicas of real-world objects or systems—are particularly useful for this. They can establish realistic, controlled settings for AI training while also connecting to real-world systems to collect more data as needed. In this study, we examined 22 current research publications to see how digital twins are being used to support AI simulations. We assessed existing trends, provided a framework for combining digital twins and AI, and highlighted major difficulties and future possibilities for researchers.
Featured Image
Photo by Steve Johnson on Unsplash
Why is it important?
Our article is the first to present a comprehensive evaluation of recent research on the use of digital twins in AI simulations, as well as a coherent framework for explaining how digital twins interact with AI components. We also discuss major trends, challenges, and future research opportunities, demonstrating how digital twins can significantly enhance AI training and lead to more accurate, trustworthy models.
Perspectives
Read the Original
This page is a summary of: AI Simulation by Digital Twins: Systematic Survey of the State of the Art and a Reference Framework, September 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3652620.3688253.
You can read the full text:
Contributors
The following have contributed to this page