Efficient Neuromorphic Vision and Perception for Long-Term Robotic Autonomy

About the Project

Autonomous robots are increasingly deployed in industries such as precision agriculture, environmental monitoring, and logistics. However, their ability to operate efficiently in real-world environments is constrained by the high computational demands and energy consumption of traditional AI-based vision systems. Conventional cameras capture redundant information, requiring heavy processing, while deep learning models used for robotic perception rely on power-hungry hardware.

Neuromorphic vision and computing offer a bio-inspired solution to these challenges. Instead of capturing full images at fixed intervals, event-based cameras detect only changes in the scene, reducing data load and enabling low-power, high-speed robotic perception. Paired with spiking neural networks (SNNs) running on neuromorphic processors, this approach allows robots to process sensory information efficiently, mimicking how biological brains work.

This PhD project explores neuromorphic vision and efficient robotic perception to enable energy-efficient, real-time autonomy in dynamic environments.

There is a wide range of possible research directions, allowing the student to focus on an area that aligns with their interests. Some examples include:

• Neuromorphic vision for robotic navigation: Using event-based cameras to improve motion estimation, obstacle detection, and depth perception in autonomous ground or aerial robots.
• Spiking neural networks for decision-making: Developing SNN-based control algorithms that enable adaptive, energy-efficient planning and real-time responses.
• Long-term autonomy for micro-robots: Investigating how neuromorphic computing can extend battery life and operational efficiency for small robots in real-world applications.
• Hybrid neuromorphic-perceptual learning systems: Combining event-based vision with other sensor modalities, such as LiDAR and inertial sensors, to create robust multimodal perception systems.
• Self-supervised learning for neuromorphic robots: Exploring ways for autonomous robots to improve perception over time without human supervision, reducing reliance on large annotated datasets.

By integrating neuromorphic sensing and processing into robotic platforms, this project aims to develop lightweight, energy-efficient, and adaptable systems that can operate autonomously for extended periods in challenging environments, paving the way for sustainable next-generation robotics.

Our Resources

Durham University hosts the UK regional supercomputer, Bede, with 128 NVIDIA V100 GPUs. In addition, our Department hosts a NVIDIA CUDA Centre that caters to the increasing GPU demands for research purposes.

Our laboratories have an array of computer science equipment and sensors, including sensor-mounted vehicles, LiDAR, RADAR, EEG, drones, robots, cameras and more.

Supervision

You will be supervised by Dr. Amir Atapour-Abarghouei (Durham University Profile), an Associate Professor in Computer Vision and Machine Learning at the Department of Computer Science, Durham University. His research spans computer vision, deep learning, robotic perception, and neuromorphic computing, with a focus on efficient AI for autonomous systems.

Dr. Atapour-Abarghouei has published in top-tier conferences and journals, including CVPR, ICCV, ECCV, ICML, IEEE Transactions on Image Processing, and IEEE Transactions on Multimedia. His work has been widely cited, demonstrating a strong impact on the fields of AI-driven vision, robotics, and machine learning. He has been involved in multiple national and international research projects, including EU-funded initiatives and collaborations with industry leaders in autonomous robotics, AI-driven perception, and deep learning efficiency.

During the PhD study, you will receive comprehensive research training, including:

• Regular one-to-one meetings to guide your research direction, ensure steady progress, and refine your problem-solving skills.
• Support in academic writing and publishing, helping you target high-impact conferences and journals.
• Collaboration opportunities within a dynamic and interdisciplinary research group, allowing you to work with experts in AI, robotics, and computational intelligence.
• Access to cutting-edge computational resources, including Durham University’s GPU cluster and state-of-the-art robotic platforms.
• Opportunities to engage with industry and external collaborators, facilitating real-world applications and potential career pathways beyond academia.

Dr. Atapour-Abarghouei has supervised numerous undergraduate, MSc, and PhD students, many of whom have gone on to pursue successful careers in research, academia, and industry. His approach to supervision emphasises independent thinking, problem-driven research, and a supportive learning environment to help students develop into well-rounded researchers with expertise in cutting-edge AI and robotics technologies.

Durham University

Durham University is a world top-100 university and is ranked the 6th in the UK. As a member university of the elite Russell Group, Durham University focuses on research excellence delivered by world-leading academics. It is the third oldest university in England, following Oxford and Cambridge, with the campus's Cathedral and Castle being a UNESCO world heritage. It is located at Durham in North East England – one of the safest cities in the UK with an affordable living cost.

Entry Requirements

• A relevant undergraduate or master's degree with good scores.
• Knowledge of modern programming languages.
• Meet Durham's English requirements (https://www.dur.ac.uk/study/international/entry-requirements/english-language-requirements/).

How to Apply

Please send an email with your resume, transcripts, and any supporting documents to Dr Amir Atapour-Abarghouei at amir.atapour-abarghouei@durham.ac.uk for an initial discussion.