This position will research, develop and demonstrate the use of spiking neural networks that enable power-constrained robots to adapt to their environment continually. The project will focus on the place recognition task, where both “slow” adaptation (urban vs rural, or day vs night) as well as “fast” adaptation (sudden onset of rain, entering tunnels) is required, potentially in environments that differ significantly from the training data. The student will get the opportunity to deploy the spiking networks on Intel’s new Loihi 2 neuromorphic processor and collaborate with Intel research scientists. QCR has a range of robots to deploy the networks on, for example, the Clearpath Jackal robot, which is equipped with a range of sensors including an event camera that is particularly well suited to work in conjunction with spiking neural networks.
Do you have a strong background in mathematics, computer science or engineering? Are you interested in computer vision and robotics, and are wondering how nature has solved complex navigation tasks?
This position is funded by Intel’s Neuromorphic Computing Lab through a grant to Dr Tobias Fischer and Professor Michael Milford. The grant provides a PhD scholarship with a significant top-up component for three years.
Location – Gardens Point
External (open to everyone)
To be filled ASAP
- your CV
- your academic papers
- your IELTS or equivalent (if applicable)