Research Supervisor Connect
This project focuses on developing digital twin systems powered by generative AI to enable realistic modelling, simulation, and prediction of complex physical and cyber-physical systems. The research aims to create adaptive, data-driven digital replicas that support optimisation, decision-making, and innovation across diverse domains.
Electrical and Computer Engineering
The project seeks to advance the next generation of digital twin technology by integrating generative artificial intelligence with real-time sensing, simulation, and predictive analytics. Traditional digital twins rely heavily on physics-based models and structured data, limiting their adaptability to dynamic environments. This research will employ generative models such as GANs, variational autoencoders (VAEs), and large foundation models to learn complex system dynamics, generate synthetic datasets, and simulate realistic scenarios for training, testing, and optimisation. The project will explore hybrid approaches that combine physics-informed modelling with AI-driven generative capabilities, enabling digital twins to self-update, adapt, and provide predictive insights in applications such as healthcare, smart manufacturing, urban systems, and autonomous robotics. The outcomes will pave the way for scalable, intelligent, and resilient digital twin ecosystems.
Research Techniques/Methodologies/Technologies
The research will employ generative AI techniques (GANs, VAEs, diffusion models), physics-informed machine learning, multimodal data fusion, and real-time data integration with IoT and sensor networks. Simulation platforms and high-performance computing environments will be used for model training and validation.
Successful candidates
Candidates should have a background in computer science, electrical/electronic engineering, data science, or related fields. Strong skills in machine learning, deep learning, and programming (Python, TensorFlow, PyTorch) are required. Experience in digital twin modelling, simulation, or domain-specific applications (e.g., healthcare, manufacturing, robotics) will be advantageous. The candidate should demonstrate strong analytical ability, creativity, and motivation for interdisciplinary research.
How to apply:
To apply, please email [email protected] the following:
The opportunity ID for this research opportunity is 3696