Machine learning based movement tracking from Internet-of-Things [IoT] devices to better understand the link between human behaviour and environmental stressors in cities

The population in cities around the world is growing. Eighty-five per cent of humanity – some nine billion people - will live in cities by the end of this century. Environmental protection, green economic growth, social equality, clean energy, sustainable mobility and health are all best addressed in cities. Digital technologies and data make new forms of governance, politics and planning possible, bringing data and monitoring logics to the fore and connecting people, governments and resources in new and potentially more efficient ways.

Smart city projects have invested heavily to establish a basis for monitoring a wide range of environmental and non-environmental variables to extract key relationships and develop smart systems. However, the effectiveness of studying one problem in isolation is becoming smaller. For example, air quality is a global problem, with regional dependencies linked to emissions, atmospheric processes and socio-economic factors. Directly monitoring air quality in every location of interest is a huge challenge. On the other hand, if we can exploit the rich data streams available to us from other platforms, we can start to build holistic monitoring platforms, and predictive techniques, that not only exploits existing datasets but also helps us inform urban design.

In this project you will use open source video capture hardware, and established object detection machine learning algorithms, to profile transport and citizen movement in cities in the UK and, perhaps, internationally. You will have the opportunity to refine these techniques whilst correlating the extracted data with environmental metrics captured from a range of smart city platforms [including air quality, weather, noise, energy use etc]. Your work will start to evaluate the potential for integrating disparate data streams in network systems to detect the impact of interventions, such as road closures, on environmental stressors. You will also have the opportunity to evaluate how transferable findings from one city are to the next. You will work in a well-established group and university. Building on our expertise in air quality, your work will benefit from ongoing efforts captured by our Data Science Institute and Urban Institute. You will sit within a hugely active area of research that promises to not only deliver impact, but embed highly transferable skills in you as an individual.