A framework for human-computer interactive street network design based on a multi-stage deep learning approach
Introduction
The accelerated urban expansion and regeneration that is driven by the increasing world population narrow the windows of time for urban planning and design (UN-Habitat, 2020). These increasing demands lead to the advancement of computer-aided planning and design tools used in the automatic or guided generation of spatial proposals, particularly in the preliminary stages of urban (re)development (Brömmelstroet, 2013; Deal et al., 2017; Yang et al., 2019). Among the various functions of computer-aided tools, street network planning and design is regarded as the backbone of generating urban spatial layouts and shaping the built environment.
A series of methods have been developed to support the planning and design of street configurations. For the conventional drawing-based design process, the concepts of “link and place” and “cooperative design” have been widely adopted to guide proposal development, which considers transport connectivity, street activities, and stakeholders' feedback (see, e.g., Jones, Marshall, & Boujenko, 2008, Laurini, 2001). For computer-based decision support tools, either rule sets or precedents are manually defined or fed into the computer systems to realize the automatic generation of urban street layouts (see, e.g., Parish & Müller, 2001).
However, user-driven design (e.g., drawing-based and procedural/rule-based approaches) overly relies on planners' disciplinary knowledge and professional experience in the quality delivery of the design proposals. The design process is led by professionals whose seniority plays a decisive role in coordinating the stakeholders' engagement. By contrast, example-driven design (e.g., learning-based approaches) focuses on the automatic generation of street configurations based on precedents, with limited consideration of users' intelligence and guidance.
To date, little attention has been given to the interactions between human and machine intelligence in the street network design process to complement the strengths of both. Among the limited relevant attempts, Fang, Jin, and Yang (2021) explored the potential incorporation of users' knowledge (e.g., design guidance) into computer models (e.g., deep learning methods) for delivering spatial proposals that can better fit within the surrounding context. Nevertheless, the method remains outcome-oriented (“click and go”), where the dynamic processes of the proposal development (i.e., the human-computer interactions) cannot be achieved.
Humans are adept at local knowledge and professional intelligence, while computers offer automatic generation and machine intelligence. To capitalize on these relative strengths for urban planning and design, this research develops a methodological framework for interactive street network design based on a multi-stage deep learning approach. The framework consists of three modules: (1) a dataset preparation module that links open-source road networks with human-labeled planning guidance; (2) a deep learning module for the automatic generation of urban road networks. The module extends the conventional end-to-end structure of deep learning techniques in street network generation (see, e.g., Fang et al., 2021) to a multi-stage model, thereby enabling the reinforcement of user-defined and iteratively developed guidance systems; and (3) a human-computer interaction (HCI) system that can facilitate the interactive and progressive design process.
The proposed framework is tested in a case study of four European cities—Amsterdam, Barcelona, Berlin, and Prague. The four cities have representative urban (re)development processes for street configurations that date from the medieval period, characterized by a rich mixture of street network patterns. Note that although the proposed framework can be universally applied to different case study areas, users are expected to follow the dataset preparation module to develop a localized dataset for model training due to varying factors that determine local street network design globally, such as history, culture, and topography.
The remainder of this paper is organized as follows. Section 2 presents a literature review on computer-aided tools for street network design. Section 3 illustrates the proposed framework for interactive street network generation. Section 4 discusses the datasets for the model training and the experiment design used for evaluating the models' performance. Section 5 presents the results of the experiment centered around the MSDL model and its ability to work within the HCI framework. Section 6 concludes the paper and discusses its contribution.
Section snippets
Literature review
Two prevalent frameworks dominate computer-aided models for street network design: procedural-based and learning-based frameworks. Procedural-based models hinge on using human-input and step-by-step rule sets to generate street layouts while learning-based models automate the street network design process by training computer models using real-world samples of street networks.
ArcGIS CityEngine is representative of a procedural-based tool, in that it is commercialized software that pipelines the
A methodological framework for human-computer interactive street network design
We present a methodological framework for an interactive street network design. The framework proposes a deep learning approach and an HCI system to complement the learning- and procedural-based computer-aided frameworks currently used in urban planning and design. As shown in Fig. 1, we develop three modules to realize the interactive design process.
The first module prepares a localized dataset on the existing street configuration as well as planning guidance for model training and testing.
Data preparation for the case study areas
Following the process introduced in Section 3.1, a localized dataset was prepared to train and test the proposed MSDL model. Four European cities (Amsterdam, Barcelona, Berlin, and Prague), featuring various street network patterns, road hierarchies, and network junctions, were selected as the areas for analysis in this case study.
To avoid model overfitting during the training stage and to guarantee that the trained model was not exposed to the data used for testing, buffer zones were created
Experiment results of the MSDL model
To visually illustrate the experiment results of the MSDL model, 12 selected pairs of data input, generation outcomes, and ground truth are shown in Fig. 5, Fig. 6, Fig. 7. The first batch of examples (Fig. 5) features limited planning guidance as input (i.e., no junction information and homogeneous street pattern type(s)). When the road network features in the context region are simple and have a strong coherence (cases E1–1 to E1–4), all four models can successfully predict street network
Discussion and conclusions
This research develops an interactive framework for street network design that complements user-driven (i.e., procedural-based tools) and example-driven (i.e., learning-based tools) approaches in urban planning and design. A multi-stage deep learning (MSDL) model is proposed to reinforce user input while the model is further incorporated into a human-computer interaction (HCI) system to enable an interactive and iterative design process.
The performance and working mechanism of the proposed
Disclosure statement
No potential conflict of interest was reported by the authors.
CRediT authorship contribution statement
Zhou Fang: Conceptualization, Methodology, Software, Data curation, Formal analysis, Writing – original draft. Jiaxin Qi: Software, Validation. Lubin Fan: Supervision, Writing – review & editing. Jianqiang Huang: Supervision, Project administration, Writing – review & editing. Ying Jin: Conceptualization, Supervision, Writing – review & editing. Tianren Yang: Conceptualization, Supervision, Writing – original draft, Writing – review & editing.
Acknowledgments
Zhou Fang appreciates the support from the China Scholarship Council and the Cambridge Commonwealth, European & International Trust through a CSC–Cambridge Trust scholarship (CSC No. 201808060082), Magdalene College via Ng Fourth-Year PhD Bursary, and the Alibaba Group through the Alibaba Research Intern Program. Ying Jin wishes to acknowledge funding support from the Tsinghua University Initiative Scientific Research Program via the Tsinghua–Cambridge research collaboration. Tianren Yang
References (35)
OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks
Computers, Environment and Urban Systems
(2017)Performance of Planning Support Systems
Computers, Environment and Urban Systems
(2013)The role of multidirectional temporal analysis in scenario planning exercises and planning support systems
Computers, Environment and Urban Systems
(2017)- et al.
Creating more people-friendly urban streets through ‘link and place’street planning and design
IATSS Research
(2008) - et al.
Interactive example-based urban layout synthesis
ACM Transactions on Graphics (ToG)
(2008) - et al.
A note on the inception score
(2018) - et al.
Procedural modelling of urban road networks
(2014) OpenStreetMap
(2010)- et al.
The graphical user interface: Crisis, danger, and opportunity
Journal of Visual Impairment & Blindness
(1990) Interactive procedural street modeling
HigherHRNet: Scale-Aware representation learning for bottom-up human pose estimation
Incorporating planning intelligence into deep learning: A planning support tool for street network design
Journal of Urban Technology
A topography-aware approach to the automatic generation of urban road networks
International Journal of Geographical Information Science
Procedural generation of roads
Computer Graphics Forum
Generative adversarial nets
Advances in Neural Information Processing Systems
Openstreetmap: User-generated street maps
IEEE Pervasive Computing
StreetGAN: Towards road network synthesis with generative adversarial networks
Cited by (4)
Computer vision applications for urban planning: A systematic review of opportunities and constraints
2024, Sustainable Cities and SocietyThe application of traditional Chinese cultural elements in urban street landscape using the Internet of Things and deep learning
2023, Journal of Intelligent and Fuzzy Systems