Procedural city generation beyond game development

Abstract

The common trend in the scientific inquiry of urban areas and their populations is to use real-world geographic and population data to understand, explain, and predict urban phenomena. We argue that this trend limits our understanding of urban areas as dealing with arbitrarily collected geographic data requires technical expertise to process; moreover, population data is often aggregated, sparsified, or anonymized for privacy reasons. We believe synthetic urban areas generated via procedural city generation, which is a technique mostly used in the gaming area, could help improve the state-of-the-art in many disciplines which study urban areas. In this paper, we describe a selection of research areas that could benefit from such synthetic urban data and show that the current research in procedurally generated cities needs to address specific issues (e.g., plausibility) to sufficiently capture real-world cities and thus take such data beyond gaming.

Urban life: a model of people and places

“…Unlike manual data generation that needs substantial human effort, procedural data generation is performed by a procedure to automatically generate content and data (Kim et al. 2018 ). Simply stated, an L-system is a string rewriting system that can be used to generate fractals with a dimension ranging from 1 to 2.…”

“…Further information about our procedural city generation and its role in agent-based modeling can be found in Kim et al. ( 2018 ). We provide several cities that differ in size for use by the research community such as those interested in exploring algorithms for location-based social networks (Kim et al.…”

“…The road networks are generated using procedural city generation based on L-systems (Lindenmayer 1968) along with considering different types of city morphologies. Unlike manual data generation that needs substantial human effort, procedural data generation is performed by a procedure to automatically generate content and data (Kim et al 2018). Simply stated, an L-system is a string rewriting system that can be used to generate fractals with a dimension ranging from 1 to 2.…”

“…Our procedurally generated world has buildings that are connected through a spatially explicit road network (Kim et al. 2018 ). Agents move across this network to find places (i.e., locations) to eat, to work, to find shelter (i.e., homes), to follow recreational activities, and to meet their friends.…”

Data-driven mobility models for COVID-19 simulation

“…These activity sequences are informed by time-use and transport surveys or by patterns extracted from phone records [35]. POIs may be randomly generated across the study area [25] or are represented in the model using geospatial data, although the former is more common. An agent's workplace, school, or visit to public POIs such as restaurants or grocery stores are either selected at random [23,24,31,32,37] or selected by a function of the distance between the agent's home location and the POI [1,16,31].…”