Neuromorphic Optical Sensors

Project ID: 2228cd1297 (You will need this ID for your application)

Under Offer

Research Theme: Information and Communication Technologies

UCL Lead department: Electronic and Electrical Engineering (EEE)

Department Website

Lead Supervisor: John Labram

Project Summary:

Why this research is important: The rapid growth of data is driving applications like self-driving cars, IoT, medical tech, and more. However, the AI hardware they depend on currently uses excessive energy. The costs of training advanced AI models are skyrocketing, similar to computing power expenses. Recently, neuromorphic or brain-inspired technologies have gained attention as potential energy-efficient computing alternatives with added capabilities. While great progress has been made in the processing of information based on neuromorphic principles, signals provided to this hardware still take a form designed for traditional von Neumann architectures. For example, conventional image sensors reproduce every aspect of the visual field, representing brighter regions of a scene as brighter regions in an image. Such signals are not well suited for neuromorphic processing. An alternative strategy is to use event-driven sensors, which output a signal only in response to changes in light intensity, not to constant light intensity.

What you will be doing: In this project, you’ll design and characterise new sensors using technologies such as memristive oxide devices, traditional semiconductors, and devices based on precisely-stacked 2D materials, such as transition metal dichalcogenides, their oxides, and graphene. You will deposit semiconductors, dielectrics, and metals, in combination with micro and nano-fabrication techniques to form the device. You will undertake a series of opto-electronic experiments on these sensors. Your objective will be to generate knowledge on their operating principles, and to develop strategies to improve their performance. You will begin with materials and single device characterisation and optimisation. Additionally, you’ll create a compact device model for simulations at higher abstraction levels, including functional circuits and ultimately system architectures. You will collaborate with teams across UCL to scale these sensors up to high-density arrays, and to develop strategies to process and interpret signals from them.