Urban comfort—thermal ease, walkability, greenery, safety, and the many sensory qualities of everyday city life—is a core planning goal. Yet comfort remains fragmented across disciplines and metrics: thermal models, walk-score proxies, and perception surveys rarely speak the same language, and a unified framework for digital planning is still missing.

Spatial Intelligence of Urban Comfort develops computational approaches to model, evaluate, and improve human-centred comfort in urban environments. The research treats comfort not as a single index but as a multidimensional, spatially explicit phenomenon that must be supported by data, interpreted across scales, and increasingly assisted by AI.

Our framework for urban comfort assessment in the era of digital planning emphasises three pillars: multidimensional analysis (integrating physical, perceptual, and behavioural dimensions); multi-source data support (imagery, surveys, environmental sensing, and urban morphology); and AI-assisted workflows that make city-wide evaluation tractable without sacrificing interpretability.

Complementary tools—including computational comfort indexing and perception survey platforms—extend this agenda from theory to practice, connecting spatial intelligence research with actionable maps and planning workflows.

Related empirical work is listed below.

城市舒适——热舒适、步行性、绿化、安全感及日常城市生活的诸多感官品质——是规划的核心目标。然而舒适仍分散在不同学科与指标之间：热环境模型、步行指数代理与感知调查很少共用同一套语言，面向数字规划的统一框架仍然缺失。

城市舒适空间智能 发展计算方法来建模、评估并改善以人为中心的城市舒适。研究将舒适视为多维、空间显式的现象，需要数据支撑、跨尺度解读，并日益由 AI 辅助。

我们面向数字规划时代的城市舒适评估提出三大支柱：多维分析（整合物理、感知与行为维度）；多源数据（影像、调查、环境传感、城市形态）；以及 AI 辅助工作流，使城市尺度评估可行且不牺牲可解释性。

配套工具——包括计算舒适指数与感知调查平台——将这一议程从理论延伸至实践，连接空间智能研究与可操作的地图及规划流程。

相关实证工作见下文。