MegaSense helps avoid air pollution

The fine particles you breathe in affect your health, and they are not easy to avoid. The more accurate and real-time the data collected on air quality, the smarter the applications that can be designed to promote residents’ health and wellbeing as well as a clean environment.

The University of Helsinki’s MegaSense programme aims to build a global observation system which produces uniform, accurate and real-time data on harmful substances found in the air. The programme strives to remove any bottlenecks associated with measurement with the help of a comprehensive measuring station network and artificial intelligence.

Combining light and heavy measuring devices

Measuring and analysing air quality and utilising the data are complicated activities that are prone to errors, and the currently available measuring techniques do not yet produce very accurate data.

Expensive measuring stations cannot be installed very densely, while the problem with inexpensive air quality sensors is that they are unable to produce data that is as accurate as that produced by the more expensive devices.

“Combining the two opens up entirely new possibilities”, says Professor Sasu Tarkoma from the Department of Computer Science, University of Helsinki. Tarkoma heads the MegaSense programme.

Ultimately, air quality analyses and related knowledge are based on an extensive observation dataset collected through the Stations Measuring Earth Surfaces and Atmosphere Relations (SMEAR) measuring network.

Portable measuring devices are light and inexpensive. When such a portable device is used alongside the expensive state-of-the-art device measuring thousands of different particles in Kumpula, the accuracy of the less costly instrument is also improved.

Now, inexpensive sensors have also been combined with a technical solution that automatically adjusts the sensors’ measuring accuracy with the help of artificial intelligence and mathematical models known as virtual sensors.

This solution, which is called an integrated model, brings the measurements of inexpensive sensors almost up to the same level of accuracy as those of fixed measuring devices.

Are fine particles in residential areas different from the air pollution found in ports?

According to Tarkoma, artificial intelligence helps elevate air quality measurements and related measuring devices to a whole new level.

“At a reasonable cost, we are able to build devices whose measuring capacity is very reliable and accurate. We can introduce wearable devices so that measuring devices are no longer tied to a particular location.

During the past two years, a hundred or so Helsinki residents have been carrying devices that have been accumulating air quality data along their daily routes at five-minute intervals. The devices used in the pilots have been connected to 4G and 5G networks, enabling real-time monitoring of air quality.”

The residents contributing to the study were recruited from many different residential areas in Helsinki. Each of these areas has its own challenges related to air quality, from the burning of small trees to busy bus and port traffic. With very local data, down to individual streets, measures affecting air quality can be targeted to the right place. What happens to air quality if a street is closed? Accurate historical data also makes air pollution predictions easier.

Applications sought through a competition

As a rule, collected and anonymised data is openly shared to make it possible to collaboratively develop applications based on it. Companies can utilise the 5G-based platform established in the process in their product development. Tarkoma also praises the City of Helsinki, which has assumed a pioneering role by making its data available and launching pilots where new technical solutions are developed, trialled and deployed on an open platform.

What kinds of services could Helsinki residents be offered by utilising this combination of data and platforms? University of Helsinki and the technology event Slush have coordinated together with the City of Helsinki, YIT and the Finnish Meteorological Institute the Venture with Air challenge, which seeks feasible ideas and solutions for utilising air quality data for the benefit of residents and the environment. The winners will be announced on November 19th 2020.

One of the already released applications is Green Paths routing tool for the Helsinki Metropolitan Area, which helps its users avoid busy streets and still reach their destinations as quickly as possible. Its recommendations are based on real-time air quality data and noise level measurements conducted throughout the city. 

Tarkoma believes we have only witnessed the initial stages of future development in the field.

Read more

MegaSense programme
News article (20 October 2020) Artificial intelligence helps to obtain increasingly accurate information on air quality
News article (24 September 2020) The new Green Path route planner finds the cycling and walking routes with the least noise and best air quality
News article (23 September 2020) Venture with Air invites you to search for solutions to air pollution problems in cities
News article (26 March 2020) University of Helsinki and the city collaborate to support wellbeing of Helsinki residents and solve global challenges
News article (11 October 2018) In future, Helsinki residents can receive real-time and detailed air quality data – the HOPE project receives large EU funding