杨俊宴 朱骁

Yang Junyan, Zhu Xiao

关键词：人工智能；城市设计；设计方法；街区形态；人机交互

Keywords：Artificial Intelligence; Urban Design; Design Method; Block Morphology; Human-Computer Interaction

摘要：

伴随信息环境和数据基础的变化，人工智能在大数据、语音与图像识别和深度学习等方面取得了发展突破，传统的城市设计技术方法也面临着升级迭代的重要契机。本文从系统阐述国内外人工智能与城市设计技术方法的结合模式入手，以城市中广泛存在并能衔接宏观城市尺度与微观建筑尺度的街区尺度为研究对象，通过进化算法、适应性算法和有监督深度学习等人工智能方法，构建了街区三维形态的智能化设计模块，并提出一套较为完整的城市街区三维形态智能化设计技术流程。在此基础上，通过结合特定具体场地，研究了人机交互下从基底搭建、方案生成到人机交互的实践探索，提出了人工智能与设计师交互式的城市设计模式。本文以街区尺度为媒介，试图解决当下人工智能城市设计在不同尺度上的生成逻辑阻塞与内在机理不明的问题，为未来数字化城市设计技术方法的迭代转型以及相关设计辅助系统的开发研究等提供参考与借鉴。

Abstract：

With the changes in the information environment and data foundation, artificial intelligence has made breakthroughs in the development of big data, language image recognition and deep learning. Traditional urban design technology methods are also facing important opportunities for upgrading and iteration. This paper takes the block scale which is widely available in the city as the research object. Through artificial intelligence methods such as evolutionary algorithms, adaptive algorithms and supervised deep learning, an intelligent design module for the three-dimensional shape of the block is constructed. On this basis, by combining specific sites, the practical exploration of human-computer interaction from base construction, plan generation to human-computer interaction is studied and an interactive urban design model of artificial intelligence and designers is proposed. Using the neighborhood scale as the medium, this paper attempts to solve the problems of generation logic blockage and unclear internal mechanism in current artificial intelligence city design at different scales. At the same time, the paper provides references for the iterative transformation of future digital urban design technology methods and the development research of related design assistance systems.

杨俊宴（通信作者），东南大学建筑学院，教授，博士生导师；东南大学智慧城市研究院，副院长。yangjy_seu@163.com

朱骁，东南大学建筑学院，博士研究生。826438993@qq.com

• [1] 吴志强. 人工智能辅助城市规划[J]. 时代建筑, 2018(1): 6-11.

  [2] 王建国. 基于人机互动的数字化城市设计——城市设计第四代范型刍议[J]. 国际城市规划, 2018, 33(1): 1-6. DOI: 10.22217/upi.2017.558.

  [3] 王建国. 从理性规划的视角看城市设计发展的四代范型[J]. 城市规划, 2018, 42(1): 9-19, 73.

  [4] 杨俊宴. 全数字化城市设计的理论范式探索[J]. 国际城市规划, 2018, 33(1): 7-21. DOI: 10.22217/upi.2017.556.

  [5] 龙瀛, 毛其智, 杨东峰, 等. 城市形态、交通能耗和环境影响集成的多智能体模型[J]. 地理学报, 2011, 66(8): 1033-1044.

  [6] LIU L, SILVA E A, WU C, et al. A machine learning-based method for the large-scale evaluation of the qualities of the urban environment[J]. Computers, environment and urban systems, 2017, 65: 113-125.

  [7] ASCHWANDEN G D, WIJNANDS J S, THOMPSON J, et al. Learning to walk: modeling transportation mode choice distribution through neural networks[J]. Environment and planning b: urban analytics and city science, 2019: 239980831986257.

  [8] GEORGIOS N K. The application of artificial intelligence in public administration for forecasting high crime risk transportation areas in urban environment[J]. Transportation research procedia, 2017, 24: 467-473.

  [9] 刘伦, 王辉. 城市研究中的计算机视觉应用进展与展望[J]. 城市规划, 2019, 43(1): 117-124.

  [10] WIJNANDS J S, NICE K A, THOMPSON J, et al. Streetscape augmentation using generative adversarial networks: insights related to health and wellbeing[J]. Sustainable cities and society, 2019, 49.

  [11] 格哈德·施密特. 人工智能在建筑与城市设计中的第二次机会[J]. 时代建筑, 2018(1): 32-37.

  [12] 李飚, 郭梓峰, 李荣“. 数字链”建筑生成的技术间隙填充[J]. 建筑学报, 2014(8): 20-25.

  [13] FILOMENA G, VERSTEGEN J A, MANLEY E. A computational approach to‘ the image of the city’[J]. Cities, 2019, 89: 14-25.

  [14] PARISH Y I H, MüLLER P. Procedural modeling of cities[C] // Processing of ACM SIGGRAPHL, 2001: 301-308.

  [15] 宋靖华. 基于生成式设计的居住区生成强排方案研究[C] // 全国高等学校建筑学专业教育指导委员会建筑数字技术教学工作委员会. 数字技术·建筑全生命周期——2018 年全国建筑院系建筑数字技术教学与研究学术研讨会论文集, 2018: 6.

  [16] 张林军, 吴志强. 居住区规划设计中计算机生态模拟软件运用的评价与优化[C] // 中国城市科学研究会. 第六届国际绿色建筑与建筑节能大会论文集, 2010: 9.

  [17] 孙澄宇, 宋小冬. 深度强化学习：高层建筑群自动布局新途径[J]. 城市规划学刊, 2019(4): 102-108.

  [18] 唐芃, 李鸿渐, 王笑, 等. 基于机器学习的传统建筑聚落历史风貌保护生成设计方法——以罗马Termini 火车站周边地块城市更新设计为例[J]. 建筑师, 2019(1): 100-105.

  [19] 孙澄宇, 罗启明, 涂鹏, 等. 街坊尺度下建筑群体三维体量的自动生成方法初探[J]. 城市建筑, 2016(1): 114-117.

  [20] OH J, HWANG J-E, SMITH S F, et al. Learning from main streetsa machine learning approach identifying neighborhood commercial districts[M] // VAN LEEUWEN J P, TIMMERMANS H J P, eds. Innovations

  in design & decision support systems in architecture and urban planning. Dordrecht: Springer, 2006: 325-340.

  [21] 黄陈瑶, 吉国华. 产业园区建筑的自动布局实验——基于遗传算法优化[J]. 安徽建筑, 2019, 26(4): 6-8.

  [22] 季惠敏, 丁沃沃. 基于量化的城市街廓空间形态分类研究[J]. 新建筑, 2019(6): 4-8.

  [23] NADERI J R, RAMAN B. Capturing impressions of pedestrian landscapes used for healing purposes with decision tree learning[J]. Landscape and urban planning, 2005, 73(2/3): 155-166.

  [24] 王建国. 包容共享、显隐互鉴、宜居可期——城市活力的历史图景和当代营造[J]. 城市规划, 2019, 43(12): 9-16.

  
  

• 收起