|£18,622 UKRI rate (for 2023/24)
|23rd November 2023
|15th February 2024
How to apply: uom.link/pgr-apply
The University of Manchester will provide fully fund this PhD; tuition fees for home students is available. We will also provide a tax free stipend set at the UKRI rate (£18,622 for 2023/24).
The Robotics for Extreme Environments Group invites applications for a PhD studentship to develop vision-based simultaneous localisation and mapping (SLAM) for usage in low-light and obstacle rich environments. The successful candidate will work at the RAICo One facility in Cumbria, where they will have the opportunity to work in close partnership with robotics and vision experts from the University of Manchester, Sellafield Ltd and the UK Atomic Energy Authority.
Equipping robots with vision capability is an active research area. It aims to enable robots to sense surroundings and detect obstacles autonomously. Challenges lie in operating in several different and challenging environment types and conditions. One such a condition is low/no light environments which can arise in many situations, for example during inspections and operations in hazardous (often after disaster) environments. Additionally, these types of environments are often cluttered with difficult to detect and avoid obstacles or objects such as cables. Of particular interest is how to perform SLAM in these conditions. This project will explore various vision-based sensors in particular lightweight, low-cost and reliable and investigate their capabilities for low-light conditions and how they can be integrated in the SLAM or detection and navigation operations. The project will also explore and develop corresponding suitable SLAM algorithms for effective operations in autonomous systems under such environments.
Successful candidate will work on solutions for the problem of vision-based navigation in limited light conditions using machine learning and/or hardware-based approaches such as embedded edge devices. Some potential solutions include event-based cameras, thermal cameras, extra physical lights, and machine learning to either enhance low-light image or learn to extract useful features from low light image.
This project will be undertaken in the RAICo One facility (hotrobotics.co.uk/facilities/university-of-manchester-2), which is located in Whitehaven, Cumbria. The facility is a collaboration between The University of Manchester, the United Kingdom Atomic Energy Authority, the Nuclear Decommissioning Authority and Sellafield Ltd. The facility is well equipped with a range of robotic systems, including a Unitree B1, several Boston Dynamic Spots and various AgileX, Clearpath, Kuka and Kinova robots. RAICo One also has dedicated robotic test areas that include unique demonstration areas for aerial, ground and aquatic robots. RAICo One focuses on developing robots for nuclear decommissioning which can operate in areas inaccessible to humans. The facility is in Cumbria, next to the Lake District, which is one of the most beautiful regions of the UK.
Type / Role: