Vector-based Pedestrian Navigation in Cities

TL;DR

This paper introduces a vector-based navigation model for pedestrians in cities, showing it predicts real-world GPS-traced paths better than shortest-distance models and suggests a universal human path planning strategy.

Contribution

It presents a novel vector-based model of pedestrian navigation that outperforms traditional shortest-path models in predicting real-world urban walking routes.

Findings

People deviate more from shortest paths as distance increases.

Paths differ significantly when origin and destination are swapped.

Vector-based model better predicts human paths than distance-minimizing models.

Abstract

How do pedestrians choose their paths within city street networks? Researchers have tried to shed light on this matter through strictly controlled experiments, but an ultimate answer based on real-world mobility data is still lacking. Here, we analyze salient features of human path planning through a statistical analysis of a massive dataset of GPS traces, which reveals that (1) people increasingly deviate from the shortest path when the distance between origin and destination increases, and (2) chosen paths are statistically different when origin and destination are swapped. We posit that direction to goal is a main driver of path planning and develop a vector-based navigation model that is a statistically better predictor of human paths than a model based on minimizing distance with stochastic effects. Our findings generalize across two major US cities with different street networks, hinting to the fact that vector-based navigation might be a universal property of human path planning.