The route to cleaner air

A three-year research project jointly funded by the Department for Transport and EPSRC is to create a sensor network that will allow traffic pollution to be reduced by managing vehicle movements in real time.

The Pervasive Mobile Environmental Sensor Grids (PMESG) study is due to be launched at the World Congress on Intelligent Transport and Systems and Services at London‘s Excel arena in October.

Transport-based grid

The 3.5m project involves a collaboration between the universities of Leeds, Cambridge, Newcastle and Southampton and Prof John Polak at ImperialCollege, London.

It aims to create four proof of concept demonstrators, each different, in London, Cambridge, Gateshead and Leicester. Industrial partners include IBM UK Labs, SERCO, Boeing and Thales Research. Local authorities and government agencies involved include Hampshire County Council, the Highways Agency and Transport for London.

The PSMEG study’s goal is to use new sensors and e-science to create the first transport-based grid that can gather, transmit, integrate, model and interpret vast quantities of highly diverse spatially and temporally varying sensor data.

The results will allow traffic engineers to minimise the impact of road traffic on local air quality and individuals’ exposure to air pollution.

Improving vehicle technologies to manage pollution requires increasingly detailed knowledge of how traffic-generated pollution behaves in the urban environment. Factors such as street and building design, vehicle braking and accelerating patterns and local weather conditions can all affect the concentration of pollutants.

The demonstrators will use a sensor called a mote, developed at NewcastleUniversity. These are able to communicate with each other and can be fixed to lamp-posts and buildings or placed within vehicles.

If a sensor is placed within a car that then passes a lamp-post that is also fitted with one, the two will communicate with each other, passing data from the vehicle to the fixed object. The information is then relayed to another sensor further along the road and eventually to a computer linked to the internet, where it is passed to a central computer for analysis, forming ‘gateways’ in a city.

The sensors are known as smart lumps owing to their size – they resemble a matchbox. However, the researchers believe they will be able to create smart dust one day as nanotechnology allows them to bring their size down.

‘In the far future they will be made from biodegradable materials,’ said Prof Margaret Bell, Professor of Traffic and Environment Pollution at LeedsUniversity‘s Institute for Transport Studies. ‘This will allow them to be scattered and work until they disintegrate.’

Real-time data

Once the PSMEG is built, instead of relying on models to provide data on noise and pollution, traffic engineers will be able to access real-time information such as where all the buses in a city are any time. This will allow them to alter traffic flows to particular areas if congestion or pollution there rises too high.

At present, if a factor changes such as the temperature in the city rising abnormally high, analysis of pollution data using models must be done offline and may take weeks to calculate. If real-time data on factors such as the number of cars on the road can be collected and analysed online, strategies such as manipulating traffic to move queues to open spaces can quickly be put in place.

‘If there is normally congestion around a school at 9am the local authority can make decisions concerning flow of traffic around the area with this in mind,’ said Bell.

‘However, if there is an accident one day then they can’t take actions such as changing signal patterns to avoid further congestion as they have no idea of the exact pattern of traffic flow around the area at the time – all they know is the average. This will allow them to be more responsive and manage traffic better, particularly local, short-lived pollution problems.’

A finalised system is unlikely to be operational at the end of the project. However, the researchers hope their demonstrators will be complete and able to provide data.

‘In 1986 the technologies for road user charging were demonstrated, yet they were only implemented just under 20 years later,’ says Bell. ‘It will probably be a decade before this is available.’