Dr Rast has been intimately involved with neural networks, and in particular, 'neuromorphic' hardware: chips that in some sense work 'like the brain' since his school days studying for A levels. At Brookes he is a Senior Lecturer and 3rd year Tutor, doing research and teaching with specialism in Artificial Intelligence. In particular he is strongly involved with autonomous robotics, event-driven cameras and perception, large-scale neural modelling, scene understanding, and probabilistic reasoning in time.
In addition to the VAIL, he is also associated with the Autonomous Driving and Intelligent Transport (ADIT) group where he collaborates on AI issues related to autonomous driving, particularly involving perception, and Simultaneous Localisation and Mapping. He is actively involved with the OBR:Autonomous racing team as a staff consultant.
Before joining Brookes he worked (in reverse order) at the University of Southampton on large-scale parallel event-driven systems, at the University of Manchester on cognitive robotics and neuromorphic chips, and at Inficom, Inc, a wireless startup company, on control processor technology and modulation theory. All of this work was closely associated with high-performance neural network processing which remains his area of focus.
Teaching and supervision
I am Third Year tutor for all computing programmes, with modules including:
- Artificial Intelligence
- Cognitive Robotics
- Autonomous Intelligent Systems
- Artificial Intelligence Systems Engineering
I am currently part of a Technology Transfer Partnership with Supponor involving the development of advanced systems for active scene infilling. I am actively involved with the ADIT group regarding research in advanced perception techniques for autonomous racing. I am involved with the Institute for Ethical AI regarding decision-making and reinforcement learning in ethical agents. I am also engaged with the International Institute of Chocolate and Cacao Tasting developing neural models and systems for fine flavour cacao and chocolate identification and assessment.
My main research work focussses on the integration of event cameras with neuromorphic cognitive processors for autonomous neurorobotics (i.e, completely integrated robotic systems with event-driven neural hardware that can sense and act in unstructured natural environments)
- Artificial Intelligence and Robotics Group (AIR)
- Autonomous Driving and Intelligent Transport
- Visual Artificial Intelligence Laboratory (VAIL)
Fursa I, Fandi E, Musat V, Culley J, Gil E, Teeti I, Bilous L, Sluis IV, Rast A, Bradley A, 'Worsening Perception: Real-time Degradation of Autonomous Vehicle Perception Performance for Simulation of Adverse Weather Conditions'
SAE International Journal of Connected and Automated Vehicles 5 (1) (2022) pp.87-100
ISSN: 2574-0741 eISSN: 2574-075XAbstract Published here Open Access on RADAR
Rast AD, Adams SV, Davidson S, Davies S, Hopkins M, Rowley A, Stokes A, Wennekers T, Furber S, Cangelosi A, 'Behavioral Learning in a Cognitive Neuromophic Robot: An Integrative Approach'
IEEE Transactions on Neural Networks and Learning Systems 29 (12) (2018) pp.6132-6144
ISSN: 2162-237XAbstract Published here
Hernández García D, Adams S, Rast A, Wennekers T, Furber S, Cangelosi A, 'Visual attention and object naming in humanoid robots using a bio-inspired spiking neural networks'
Robotics and Autonomous Systems 104 (2018) pp.56-71
ISSN: 0921-8890Published here