The QUT Centre for Robotics has a long history in bio-inspired robotic localisation and navigation. We have recently received a grant from Intel’s Neuromorphic Computing Lab to explore the capabilities of Intel’s Loihi 2 neuromorphic research chip. Led by Dr Tobias Fischer, Professor Michael Milford, and PhD students Somayeh Hussaini and Gokul Nair, the QUT research team will investigate the use of spiking neural networks (SNNs) to enable simultaneous localisation and mapping (SLAM) algorithms in unknown or GPS-deprived environments.
The team will focus on Visual Place Recognition, a particularly challenging aspect of SLAM, which requires robust recognition and discrimination of hundreds of thousands of locations in different conditions. Drawing inspiration from the animal kingdom, the QUT team will develop biologically-inspired navigation and map formation algorithms that surpass today’s state-of-the-art in the field. The project will explore the exciting new possibilities of neuromorphic computing in the field of robotics and the potential of Loihi 2 to support difficult optimisation problems.
Funding / Grants
- Intel Labs (2022 - 2024)
- Hussaini, Somayeh, Milford, Michael, Fischer, Tobias (2022) Spiking Neural Networks for Visual Place Recognition Via Weighted Neuronal Assignments IEEE Robotics and Automation Letters, 7 (2), pp.4094-4101.
- Fischer, Tobias, Milford, Michael (2022) How Many Events Do You Need? Event-Based Visual Place Recognition Using Sparse But Varying Pixels IEEE Robotics and Automation Letters, 7 (4), pp.12275-12282.
- Fischer, Tobias, Milford, Michael (2020) Event-Based Visual Place Recognition With Ensembles of Temporal Windows IEEE Robotics and Automation Letters, 5 (4), pp.Article number: 9201344 6924-6931.
- Somayeh Hussaini, Michael Milford, Tobias Fischer (2023) Ensembles of Compact, Region-specific & Regularized Spiking Neural Networks for Scalable Place Recognition. IEEE International Conference on Robotics and Automation
- Milford, Michael, Wyeth, Gordon (2008) Mapping a suburb with a single camera using a biologically inspired SLAM system IEEE Transactions on Robotics, 24 (5), pp.1038-1053.