Cities around the world are densifying. This puts pressure on urban public spaces, which are crucial for healthy cities as they contribute to peoples’ health and well-being. Urban planning practice strives to provide citizens with positive experiences in public spaces. As a design and decision-support tool for these authorities, urban-scale digital twins have emerged to test future urban-design scenarios. However, the status quo of urban-scale digital twins is full of 3D buildings, but lack humans. That is, they do not incorporate virtual citizens, i.e. software agents, that can intrinsically experience the digital environment and react to it, as real citizens would do. Incorporating agents in digital twins, would allow for testing how these citizens respond to possible urban design scenarios and how this affects their health and well-being. 

Therefore, our aim is two-fold: (i) to devise an agent architecture of intelligent and emotional citizens who can experience the city and articulate their sense of wellbeing in a digital twin, (ii) to form an interdisciplinary research community to tackle this.

An “agent architecture” is a term from computer science for a blueprint of software agents, depicting the arrangement of agent states and behaviors relevant to the setting the agent will be placed in (d’Inverno & Luck, 2004). It serves as the starting point for a modeler/software engineer to implement agents, in a simulation model or digital twin. The term “experience” refers to the mental changes that occur when humans engage in a situation that provides physical, emotional, spiritual, and intellectual values, an important indicator of citizens’ health and subjective wellbeing (Dane, Borgers, & Feng, 2019; Pine & Gilmore, 1998; Weijs-Perrée et al., 2019). Momentary experiences are the on-site and real-time experiences, the interaction process between the agent and its physical and social environment triggering emotional changes. We will focus on stress reactions/reductions because a relation with the environmental setting is established. The spatial extent of the digital twin are Pedestrian-Priority Spaces (PPS), as these are public spaces for citizens of all ages and abilities that invite them to stay and spend time, providing ample momentary experiences and opportunities for social interaction. 

• dr. Judith Verstegen
• Letícia Marçal Russo

• Gamze Dane (TU/e)
• Theo Arentze (TU/e)
• Arend Ligtenberg (WUR)

Multiple sustainable development goals (SDGs) identified by the United Nations relate directly to healthy and safe urban environments. To realize the designed targets, the appearance of streetscapes plays a vital role. Urban environments perceived as safe, pleasant, and walkable stimulate sustainable and healthy human behavior in terms of mobility, outdoor activities, access to environmental resources etc. Data on people’s environmental perception are, however, difficult to collect typically relying on a few in-situ neighborhood audits which are impossible to scale-up while being time-consuming, costly, and labor-intensive. A solution to this challenge may arise through urban big data analytics based on artificial intelligence (AI).

We propose an innovative AI-based approach to map human perceptions of streetscapes (e.g., safety, pleasantness, walkability) through a state-of-the-art deep learning computer vision-based model trained using crowdsourced stated preferences of street view data. Based on a coordinated effort exploiting synergies between three disciplines within the Faculty of Geosciences, the project will deliver robust evidence on the perceived qualities of streetscapes putting the FAIR principles into practice. We envision these data will be helpful in formulating actionable progress by making suggestions for expanding and refining SDG indicators on sustainable communities.

• drs. Maarten Zeylmans Van Emmichoven
• dr. Matthew Danish
• dr. Britta Ricker
• dr. S.M. Labib

Cities in the global South are rapidly growing in size, but many marginalised and vulnerable residents (such as lower-income households, older adults, women and people with disabilities) do not have affordable, safe and accessible public transport, which reduces their ability to have decent work, healthcare and social life. Transport planning largely ignores access inequalities but prioritises efficiency and economic benefits. This project will go beyond traditional engineering approaches by taking a novel, user-centred intersectional approach that recognises how multiple forms of discrimination (e.g. classism, sexism, ageism and ableism) intersect to produce urban mobility inequalities for marginalised groups. The central objective is to develop evidence-based insights for affordable, safe and accessible urban mobility. More specifically, we aim to: 1) explore how physical and social barriers to urban transport are widened by the existing systems and the social and economic implications of such barriers (SDG-11&9), 2) develop and contextualise measures to improve access to work (SDG-8), healthcare (SDG-3) and social life (SDG-10) through improvements in the public transport system, and 3) co-design an inclusive urban mobility evaluative framework that can provide guidelines for inclusive cities. We will apply an innovative multi-sited mixed-methods approach combining visual surveys, GPS-led-geo-narratives and multi-stakeholder hackathons. Inequalities of urban mobility will be studied in Delhi, Bengaluru and Dhaka, as these cities are experiencing major infrastructural changes and have populations with multiple access disparities. Inclusive cities with affordable, safe and accessible low-carbon public transport lead to a reduction of emissions and improvements in public health and wellbeing.

• prof. dr. Annelies Zoomers
• prof. dr. Ajay Bailey
• prof. dr. ir. Dick Ettema
• Seama Mowri

• Dr Sobin George
• Prof. Sangamitra Roy
• Dr Anindita Datta
• Dr Md. Musleh Uddin Hasan
• Dr Shanawez Hossain

Exposure to natural outdoor environments including greenery is vital for people’s health. Remote sensing via satellites is most commonly used to generate measures of greenery, however this does not capture the street-level perspective that people experience. On-site visits are also biased through subjective ratings, labor-intensive, time-consuming, and inefficient at a large-scale. We will address this research problem with an assemblage of IT methods to derive measures of greenery from street view services on the web which allows virtual navigation through urban spaces composed of geo-tagged street-level images. We will develop methods to automatically compute objective and accurate measures of greenery employing big street view data and deep learning.

• dr. Raoul Schram MSc
• dr. ir. R.W. (Ronald) Poppe
• prof. dr. Daniel Oberski