Our Research

Energy-efficient edge AI

The goal of this research pillar is to systematically address two fundamental research questions:

  • How can we identify the energy bottlenecks and optimize the energy efficiency of on-device deep learning for diverse edge devices?
  • How environmentally sustainable is an edge AI system itself?

To address above questions, our approach covers design, measurement, analysis, and benchmark of deep learning on mobile and edge devices.

[Read more…]

Mobile Augmented/Virtual Reality (AR/VR)

The goal of this research pillar is to improve system performance, especially latency and energy efficiency, of mobile AR/VR devices. We primarily focus on adapting the system configurations of mobile AR devices to address the trade-offs between user preferences and performance requirements.

We are also recently interested in prototyping systems for emerging AR applications, including:

  • Collaborative AR
  • Mobile AR with generative AI
  • AR for scientific data exploration

[Read more…]

Digital Twins for Future Mobility

The goal of this research pillar is to conduct pioneering study on digital twins for connected and automated vehicles. Our vision is to create fair, affordable, and efficient mobility solutions by leveraging digital twins and edge computing.

We are primarily interested in:

  • Visualization of mobility digital twin with NeRF
  • Cooperative perception

[Read more…]

Academia Research Partners