London air quality

Developing machine learning algorithms and data science platforms to understand and improve air quality over London


Air quality in London has improved in recent years as a result of policies to reduce emissions, primarily from road transport, however further improvements are critical to public health. By utilising city-wide air quality sensors this project is developing machine learning algorithms and data science platforms to understand and improve air quality over London.

Explaining the science

Air quality in London has improved in recent years as a result of policies to reduce emissions, primarily from road transport. However, significant areas still exceed NO2 EU Limit Values. Poor air quality has particularly been identified as a threat to health, with an estimated 9,000+ Londoners dying early every year. Similar issues affect most cities across the UK and Europe.

At the same time, a revolution is happening in air quality monitoring. Traditionally, a relatively small number of reference quality sensors are used - followed by a period of modelling to create a London-wide snapshot (currently based on 2013 data). With the proliferation of increasingly affordable air quality sensors, it is possible to monitor air pollution at thousands of different locations in a city - greatly enhancing our ability to target and prioritise planned interventions. Increasingly companies, non-profit organisations, community groups and individuals also want to monitor the air and are investing in sensors.

This project develops machine learning algorithms, data science platforms and statistical methodology to integrate data and air pollution measurements from various heterogeneous sources in order to better estimate and accurately forecast air pollution across the city of London. Given these hyper-local estimates and associated uncertainty the group then develops algorithms and optimisation techniques to inform citizens and help design and evaluate government policy.

Running routes air pollution
48 hour predictions of air quality (NO2) Central London and a running route that changes shape to minimise air pollution.


Project aims

The main challenges of this work are:

  • To ensure that data from a wide range of networks can be brought together to a single place for analysis
  • To bring data into air quality models from a range of quality of sensors
  • To ensure that we monitor the effectiveness of the different interventions planned across London
  • To present the best estimates and forecasts in a way that app and web developers can then use to inform Londoners
  • To accurately find low pollution routes for Londoners to follow when walking, cycling or running through the city

The project researchers are developing machine learning algorithms, statistical methodology and data science platforms to understand and improve air quality over the city of London. Integrating varying-fidelity heterogeneous sensors in an overall real-time monitoring network for air quality, the project will develop state of the art machine learning models for high resolution air quality forecasting and change-point detection. This will help establish the most effective places to site future sensors, and inform policy to make targeted interventions that reduce the levels of pollution in key areas and at key times.


The goals of the project will be complemented by the parallel development of APIs and mobile apps to provide reliable, frequently updated and highly localised air quality data and forecasts for Londoners. Graph optimisation algorithms will be developed to use the air quality forecasts to find less polluted routes for people walking, running and cycling around London’s streets. Algorithms will be analysed for their complexity, efficiency and practicality.

Recent updates

February 2020

  • Dr Theo Damoulas is the Insight Speaker at the BCS-IET Turing talk 2020. Theo will explain how our urban digital twin for air quality. The talk features an amazing video created by James Walsh which dives into our digital twin of London. Established by BCS and the IET, the Turing Talk honours and recognises Alan Turing's outstanding contribution to computing.
  • Virginia Aglietti is offered an internship at Microsoft.

January 2020

  • Juan Maronas joins as a visiting researcher for four months based at the Alan Turing Institute. Juan joins from the Pattern Recognition and Human Language Technology Research Center in Valencia.
  • Dr Neil Dhir joins the London air quality project at the Alan Turing Institute.
  • Virginia Aglietti has paper accepted to AISTATS 2020. The paper is titled 'Causal Bayesian Optimization' and is co-authored with Gonzalez J. & Lu X.
  • Dr Theo Damoulas gives talk at the Data Science in Transport conference. The talk was organized by the Department for Transport.

December 2019

  • Patrick O'Hara to give talk and the Walking & Cycling Innovations Conference 2019. Patrick will give a 10 minute talk at the conference in Manchester describing how machine learning and graph optimisation algorithms can be used to minimise the air pollution exposure of people moving around a city.
  • Dr Kangrui Wang's proposal to AI UK succeeds. Kangrui will demonstrate a disease-symptom diagnosis method. The event will be held on the 24-25 March 2020.
Virginia Aglietti NeurIPS 2019
Virginia Aglietti presents Structured Variational Inference in Continuous Cox Process Models at NeurIPS 2019
Ollie Hamelijnck
Ollie Hamelijnck presents Multi-resolution Multi-task Gaussian Processes at NeurIPS 2019

September 2019

August 2019

  • Congratulations to Dr Daniel Tait for the successful defence of his PhD thesis. Daniel completed his PhD at the University of Edinburgh investigating latent force models with multiplicative interactions.

July 2019

Jeremias Knoblauch Generalized Variational Inference
Jeremias Knoblauch delivers the Generalized Variational Inference workshop.

June 2019

May 2019

April 2019

  • Dr Theo Damoulas gives interview to the BBC titled Tracking the toxic air that's killing millions
  • Patrick O'Hara gives interview to Wired magazine. He discusses why the design of algorithms for urban walking, running and cycling routes which minimise air pollution is a complex challenge.
  • Patrick O'Hara receives a research assistant position at the Turing. He will be researching graph optimisation algorithms and reinforcement learning with applications to air quality.
  • Virginia Aglietti accepts an internship offer from Amazon. The internship is for three months from August 2019 until November 2019.

March 2019

  • Dr Deniz Akyildiz successfully defends his PhD thesis. He completed his PhD under the supervision of Joaquin Miguez within the Signal Processing Group at Carlos III University of Madrid.
  • Ollie Hamelijnck receives a PhD fellowship from The Alan Turing Institute. The fellowship covers a tax-free stipend of £20,500 per annum, a travel allowance and conference fund, and tuition fees for a period of 3.5 years.
  • Jeremias Knoblauch receives an internship offer from Amazon.
  • Ollie Hamelijnck wins a prestigious PhD Feuer International Scholarship in AI. The scholarship covers tuition fees for 3 years; a stipend at RCUK rate plus a top up of £2,000 per annum; and support for travel, equipment and research.
  • Dr Theo Damoulas invited to give talk in Seoul, Korea at the International Biometric Conference (IBC 2020) in July 2020.

February 2019

  • Turing chooses the London air quality project as one of five impact stories. The story is titled: Understanding urban air quality.
  • Dr Deniz Akyildiz joins the London air quality project. Deniz is a research fellow at the University of Warwick with a joint appointment between the Dept. of Computer Science and Dept. of Statistics. He recently completed his PhD at Carlos III University of Madrid.

January 2019

  • Jeremias Knoblauch has been selected as one of 21 PhD students worldwide to receive the Facebook Fellowship award. The award consists of a stipend worth $84,000 and additionally covers two years of university tuition fees.
  • Paper to appear: Aglietti. V., Damoulas. T. & Bonilla. E. (2019). Efficient Inference in Multi-task Cox Process Models. The 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019).

December 2018

November 2018

October 2018

September 2018

June 2018

May 2018

October 2017

  • The Turing's programme for data-centric engineering announced a collaboration with the Mayor of London to tackle air pollution in London using data sensors. Read articles in City AM, Business Daily and Air Quality News. Read the Turing press release.


Researchers and collaborators

Sueda Ciftci

Research Assistant/Research Software Engineer, University of Warwick

Contact info

[email protected]