Origin-Conditional Trajectory Encoding: Measuring Urban Configurational Asymmetries through Neural Decomposition

TL;DR

The paper addresses the fragmentation of urban AI by proposing origin-conditional trajectory encoding that jointly learns spatial and movement representations while preserving origin-dependent asymmetries. It uses learnable origin embeddings, a BiLSTM with conditional attention, and a shared/specific representation decomposition learned through a four-part objective to quantify cognitive asymmetries via origin-divergence and global-asymmetry metrics. Validation on synthetic cities and Beijing demonstrates that urban morphology shapes systematic configurational inequalities, and the method yields interpretable embeddings and real-world insights. The framework offers urban planners and designers quantitative tools to assess experiential equity and to design layouts with more equitable navigational experiences, as well as enabling origin-aware analytics for navigation systems.

Abstract

Urban analytics increasingly relies on AI-driven trajectory analysis, yet current approaches suffer from methodological fragmentation: trajectory learning captures movement patterns but ignores spatial context, while spatial embedding methods encode street networks but miss temporal dynamics. Three gaps persist: (1) lack of joint training that integrates spatial and temporal representations, (2) origin-agnostic treatment that ignores directional asymmetries in navigation ($A \to B \ne B \to A$), and (3) over-reliance on auxiliary data (POIs, imagery) rather than fundamental geometric properties of urban space. We introduce a conditional trajectory encoder that jointly learns spatial and movement representations while preserving origin-dependent asymmetries using geometric features. This framework decomposes urban navigation into shared cognitive patterns and origin-specific spatial narratives, enabling quantitative measurement of cognitive asymmetries across starting locations. Our bidirectional LSTM processes visibility ratio and curvature features conditioned on learnable origin embeddings, decomposing representations into shared urban patterns and origin-specific signatures through contrastive learning. Results from six synthetic cities and real-world validation on Beijing's Xicheng District demonstrate that urban morphology creates systematic cognitive inequalities. This provides urban planners quantitative tools for assessing experiential equity, offers architects insights into layout decisions' cognitive impacts, and enables origin-aware analytics for navigation systems.

Figures (1)

• Figure 1: Learning and Visualization of Urban Spatial Representations based on Conditional Encoding