Making the city smarter

City authorities in congested urban areas are turning to sensors, AI and new digital technologies for answers to the problem of air pollution. Envirotec rounds up some of the latest developments.

Man holding phone in front of tall building
A tall order? There have been technology and funding obstacles to the realisation of large-scale, city-wide air monitoring networks.

In December, London mayor Sadiq Khan announced the trial of a £750k “cutting-edge sensor air quality monitoring system”, plans revealed concurrently with the announcement that Khan is to chair a new network of global cities cooperating to tackle air pollution. The new system will dramatically boost London’s air quality monitoring network from its current level of 100 or so sites to up to 1,000.

The initiative will also begin to use mobile devices to capture relevant data, which means data will potentially be available from hundreds of thousands of locations across the city.

Khan says he wants to use the project to strengthen understanding of some of the most dangerous pollutants, including PM2.5 particles, and identify areas to prioritise with targeted measures to lower pollution levels. These could include pedestrianising roads around schools, encouraging more walking and cycling, and traffic restrictions.

A procurement process was said to be starting in December, and the Greater London Authority (GLA) had yet to decide which specific sensor technologies and mobile monitoring techniques were to be used as part of the pilot, but delivery is expected in 2018.

Cutting edge?

The deployment of such an extensive network of sensors presents a number of challenges, and seems to push at the edges of what’s possible with current technology. Air monitoring equipment tends to be large, expensive, and in need of ongoing calibration. Drift is a problem with such sensors, becoming more pronounced with variations in things like temperature and humidity. Many technologies also suffer from cross-sensitivity. For example, electrochemical NO2 sensors tend to take erroneous readings in the presence of ozone. Sensitivity has also tended to be a problem, since much of the available technology lacks the sensitivity to measure many pollutants at the levels they occur in ambient air (requiring parts per billion levels of accuracy).

Sensor shortfall

A recent blog by air quality monitoring equipment developer Aeroqual profiled some of the areas where advances are needed. For example, at present, said the firm, there are no SO2 sensors on the market that can measure reliably below 30 ppb – a bugbear since SO2 levels in most cities are now below 10 ppb. “So SO2 sensor technology is just not at a point where they can provide useful data in a general urban environment.”

Other important directions for improvement highlighted by Aeroqual include the compensation mechanisms which are used to correct for some of the weaknesses of many sensors. In some cases these are clever algorithms. Aeroqual also cites a technique of its own, where it uses a highly selective O3 sensor to correct for ozone interference on an electrochemical NO2 sensor (all NO2 electrochemical sensors being cross-sensitive to O3 and vice versa).

Another direction of improvement is with instrument design. For example, to ensure the heat build-up from any electronics does not increase the likelihood of sensor drift.

A number of firms say they are addressing these challenges. At the Consumer Electronics Show (CES) taking place in January 2018 in Las Vegas, Californian firm AerNos was expected to announce a “nano gas sensor” technology, offering significant advantages over gas sensors currently on the market, in terms of size, power, and affordability. It is also capable of measuring in the ppb range.

The start-up, founded in 2016, announced in March 2017 that it had developed a nanoscale gas sensor specifically for measuring air quality, and which employed novel materials and nanotechnology. The technology has been called “AerCNT”, the name seemingly a nod to its use of carbon nanotubes. But it also employs machine learning and data science to arrive at a solution that is able to measure multiple pollutants, while being small and affordable enough to be incorporated in mobile devices and wearables. It also looks like it could be an enabler for Internet-of-things (IoT) devices for smart city and smart home initiatives.

The firm was expected to introduce its inaugural nano gas sensor product lines at CES, positioning itself as a developer of application-specific solutions for wearable, mobile, IoT, smart home and smart city products. CEO Sundip R Doshi said the products the firm plans to announce “will disrupt multiple industries and create new product opportunities for our customers.”

A microchip on a finger
Start-up AerNos is offering a multi-pollutant air quality sensor capability on a 3mm x 3mm MEMS (microelectromechanical systems) chip.

Smart infrastructure

An important ingredient in the aspiration to introduce large-scale, city-wide air monitoring is the underlying infrastructure of connection points and connectivity. So-called “smart cities” will reduce the cost of implementing large-scale sensor networks through the provision of a ready-made power and communication infrastructure, something that already appears to be gathering pace around the world.

In its blog recently, Aeroqual suggested that smart lighting systems, for example, provide an ideal mounting point for air quality sensors. As well as supplying power, these connection points can be installed at a height to keep them out of reach of vandals while being low enough to measure the air that people are breathing at pedestrian level.

The communication links being discussed in relation to proposed smart cities would also support these sensors. For example, low cost wireless networks based on standards like Zigbee are “a highly efficient method of communication – much cheaper than using the 4G cellular network for example.”

Recent months have seen a series of announcements that suggest a growing belief in the promise of smart cities – defined by the World Bank as cities that are “technology intensive”. And January’s Consumer Electronics Show is to include a dedicated conference program on the topic, with a focus on the discussion around public safety, and how such initiatives can make inroads into challenges like air pollution.

In November an investment firm associated with Bill Gates, Mt. Lemmon Holdings, put forward $80 million to contribute to the construction of a 24,800-acre smart city development in Buckeye, Arizona. Plans for the development say it will feature high-speed public wifi, self-driving cars, and high-tech manufacturing facilities.

So far, most projects appearing under this heading have been for demonstration purposes, or to advance research. And there have been a few addressing the issue of air quality specifically. One of these is IBM’s Green Horizon project, which aims to help the city of Beijing with its air pollution, and to improve forecasting in this respect. Drawing on IoT and AI technology, the project seemingly relies on gathering huge amounts of data from sensors around the city, about things like traffic levels, wind patterns and humidity, which can then be used for predictive analysis.

IBM launched the project in 2014 and says it has since been able to generate high-resolution pollution forecasts across 1-by-1 kilometre grid squares, providing citizens with 72 hours advance warning of pollution flare ups.

Vehicle-to-infrastructure communication

In August 2017, a project called ACCRA was announced in Leeds, which has the aim of giving local authorities more power to control city vehicle emissions. The project is a collaboration between Leeds City Council, Cenex, the Transport Systems Catapult, Earthsense, Dynniq, and Tevva Motors. It aims to demonstrate smart city technology applications to support real-time emissions control, using live air-quality data to trigger hybrid engines to automatically switch to zero-emission running when driving through highly polluted areas.

Known as “active geofencing” the technology concept will be tested on a hybrid vehicle interface developed by Tevva Motors. Transportation network systems developer Dynniq will develop a decision-making engine capable of taking inputs from a range of city data, such as live air quality information and real-time traffic conditions. EarthSense will be responsible for monitoring and uploading updated local air quality levels to the interface, which will be used to trigger on-demand zero-emissions running instructions in the participating Tevva vehicles.

Paul Bate, Principal Technologist at the Transport Systems Catapult, which is leading the consortium, commented at the time:  “Vehicle to Infrastructure (V2I) communication has the promise to transform how cities manage urban traffic control and air quality regulation.”

Cenex and the Transport Systems Catapult will evaluate the application, markets, business models and scalability of the system in hopes of using the technology more widely in Leeds and potential UK Clean Air Zones.

At this stage it is difficult to evaluate the usefulness of the air quality data being generated by smart city demonstration projects, since no data has been made publicly available in this respect. However, London has said it will share its results from the forthcoming trial of large-scale deployment of air monitoring, with the possibility that something similar could be rolled out in Bengaluru, Delhi and other cities tackling toxic air (and which are participants in the the C40 Climate Leadership Group, an alliance of cities dealing with climate change issues, which is coordinating the initiative).

Funding question

The smart city concept is still contentious, with it being an open question at this stage as to who will likely have the most powerful stake in their development, whether private sector, government or something in between. Bas Boorsma, author of a new book* on smart cities, argues that the success of the concept depends upon its ability to deliver its benefits to everyone, and that a minimum level of public control should be maintained in a smart city. The book suggests public-private partnerships (PPPs) – a union between government and the private sector – as a funding mechanism. This would, he suggests, help make the development of this kind of infrastructure attractive to technology firms, who currently have little to attract them to city hall.

* A New Digital Deal (subtitled “Beyond Smart Cities. How to Best Leverage Digitalization for the Benefit of our Communities”) by Bas Boorsma.