During the Covid pandemic two undergraduate engineering students at Christ鈥檚 鈥 Andrew Wang and Mihai Ilas 鈥 worked on Project Odysseus which provided data about the busyness of London streets to the Greater London Authority (GLA) and Transport for London (TfL).
A paper published in The Computer Journal reports that, as a result of the data generated by machine learning algorithms, in the first wave of the pandemic TfL made more than 700 interventions such as increased signage and pedestrian zoning to create more space.
On Brixton high street, for example, the artificial intelligence (AI) tool demonstrated that there was overcrowding on pavements particularly near bus stops. As a result, TfL widened the pavement and moved a bus stop.
Machine learning is a branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn.
An initial project at the Alan Turing Institute - whose Chief Scientist is Christ鈥檚 Fellow, Professor Mark Girolami - monitored air quality. But when the Covid pandemic began and social distancing measures were recommended, the study was repurposed. Instead of just vehicles, Project Odysseus also monitored pedestrian activity.
The team used live traffic feeds from 900 CCTV cameras to monitor near real-time activity on the London streets during lockdown by estimating pedestrian and vehicle density.
The low-resolution footage meant that no individuals were identifiable, but algorithms calculated the distances between pedestrians.
GLA and TfL used the data generated to understand how policy changes on social distancing - the 鈥榯wo metre鈥 rule - affected people鈥檚 behaviour and the busyness of the streets and were able to make effective interventions to create space.
Mr Wang, a data scientist in industry, said:
鈥淭his was an exciting project and I鈥檓 grateful for the opportunity to have worked on pressing public problems with leading researchers at the Alan Turing Institute.
The chance to be involved came up during College supervisions with Professor Girolami and I encourage Christ鈥檚 students to always be open to new opportunities to build relationships with world-leading academics.鈥
Mr Ilas, now a software engineer, said:
鈥淚t was great to translate the knowledge from our courses into a project where we enjoyed a lot of responsibility and freedom of experimentation at the same time, and to make an impact during unusual times.鈥
Project Odysseus is a collaboration between researchers at the University of Warwick and the Alan Turing Institute, the UK鈥檚 national institute for data science and AI.
James Walsh, Oluwafunmilola Kesa, Andrew Wang, Mihai Ilas, Patrick O鈥橦ara, Oscar Giles, Neil Dhir, Mark Girolami, Theodoros Damoulas, 鈥樷, The Computer Journal (March 2023)
Announcement: March 2024
This paper is the winner the . The award is presented once a year to the authors of the best paper published in the volume of The Computer Journal from the previous year, based on originality and quality of theme and treatment.