PhD studentships in Artificial Intelligence

Artificial Intelligence Grand Challenge

The London NERC DTP is offering two Industrial (CASE) studentships to start in September 2018. These are funded through the NERC National Productivity Investment Fund for work on projects with industrial partners that underpin growth, productivity and rebalancing of the national economy, within the context of excellence in the environmental sciences. These studentships are directly aligned to the Industrial Strategy Grand Challenge of Artificial Intelligence & Data-Driven Economy to put the UK at the forefront of the industries of the future.

The following specific projects are available, or applicants can propose an alternative project matched to the requirements of the NPIF strategy. Applicants should indicate which project(s) they are interested in, on their application.

Comprehensive mapping of tree species for the United Kingdom.

Brief description: Currently the tree coverage and species map of the UK is compiled from survey data at a very coarse spatial resolution.  Rezatec Ltd, in a pilot study with DEFRA, has demonstrated that machine learning techniques can be applied to earth observation data to map tree species effectively. This is especially valuable when studying woodland plant health where species-specific outbreaks of pests and diseases need to be contained. Meanwhile aerial imagery and LIDAR have been seen to be effective at identifying individual trees, but not their characteristics. This project will investigate a holistic approach to mapping trees, considering all the available data sets and developing new methods to characterise individual trees, within and beyond woodlands. This project would enable DEFRA to monitor all trees in the UK and mitigate against major outbreaks of pests and diseases. It aims to expand the R&D capability of a rapidly expanding UK SME. The world-leading technology developed in this project can be exported and applied to many other forested regions of the world, and may lead to patents applications. It will also provide an essential tool for DEFRA when planning post-Brexit rural strategies and funding and aligns with Governmental Industrial and Environmental strategies.

CASE partner: Dr Iain Williams, Deputy Chief Scientific Adviser at DEFRA, Dr Andy Carrel at Rezatec Limited (SME based in the Satellite Applications Catapult, Harwell) and UK local authorities

Main supervisor: Profs Maslin and Lewis (Geography, UCL)

Reducing the environmental impact of public transport networks using advanced machine learning techniques.

Brief description: This project will exploit the world-leading applied AI and ML expertise developed in the field of particle physics and astronomy to reduce the environmental impact reliability of London’s transport system, building upon successful joint projects already underway. The project will apply cutting-edge machine learning tools such as recurrent and deep learning NNs, to time series data from TfL systems, to reduce failures in service, optimise the use of the network, and thus reduce environmental impact and carbon emissions. This will have significant impact, enhancing the reliability and performance of the networks.

CASE partner: Transport for London

Main supervisor: Dr Tim Scanlon (Dept of Physics & Astronomy, UCL)

Application of Machine Learning to Earth Observation Data Analysis within an African Crop Pest Risk Information Service.

Brief description: An estimated 40% of the world’s crops are lost to pests, with Africa unduly affected, so a 5-year multi-million pound project funded by the International Partnership Program of the UK Space Agency is currently working to create a Pest Risk Information Service (PRISE) to aid Africa farming. This system will use Earth Observation, meteorological and in situ data to provide real-time pest risk forecasts and early warnings to smallholder farmers, with the aim to deliver accurate, timely information sufficient to take preventive action and increase resilience to pest outbreaks. The core of the PRISE system is a series of biological models of pest development driven by landscape and air temperature information derived largely for satellite Earth Observation. This PhD will explore the potential utilization of machine learning within these aspects of PRISE, and will test whether particular aspects of the system built currently on bio-physical and statistical modelling maybe better represented by machine learning. The project is a case studentship between King’s College London and Assimila Ltd and the student will also receive regular input from CABI who are the partners with responsibility for the biological models of pest development. In addition to satellite remote sensing and modelling, fieldwork and travel to a number of African countries working with PRISE will be a component of this PhD.

CASE partner: Drs Zofia Stott and Bethan Perkins at Assimila Ltd (UK SME in Reading).

Main supervisor: Prof. Martin Wooster (Dept of Geography, King’s College London)

Alternative Projects

Applicants may apply to work on an alternative project, if they have discussed the project with a potential supervisor in advance of applying, and have that supervisor’s support for their application. The project must be within the NPIF and NERC remit, and focussing on the Artificial Intelligence & Data industrial strategy.

Details of the project proposal and contact information for the supporting supervisor should be included in the covering letter of the studentship application.

The successful applicants will join Cohort 5 of the London NERC DTP in September 2018 and will participate alongside the rest of the cohort in selected training opportunities in their first year, including Induction Week, Skills Fridays at the Natural History Museum and the 10 day California fieldtrip.

Eligibility

This funding is for UK and EU residents only, but under exceptional circumstances other candidates may be considered. See Eligibility Criteria for full details.

Apply

Apply through our website  Applications are being considered on a rolling basis, until all positions are filled.

Interviews

Shortlisted candidates will be invited to interview.

Tagged with:

Stay informed

Click here to subscribe to our RSS newsletter by email.


Find Us

University College London is the administrative lead.

Pearson Building, UCL, Gower Street, London, WC1E 6BT

Follow us on Twitter