AccessMap Website Demonstration: Individualized, Accessible Pedestrian Trip Planning at Scale
Nicholas Bolten, Anat Caspi · 2019 · Proceedings of the 21st International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS) · doi:10.1145/3308561.3354598
Summary
This demonstration paper presents AccessMap, an open-source, city-scale interactive web map that provides individualized pedestrian infrastructure information and automatic trip planning for people with diverse mobility needs. The core problem is that traditional maps and navigation services treat pedestrians as a monolithic group — a simple extension of vehicle-centric routing — ignoring the significant heterogeneity in pedestrian mobility requirements. Roughly 15-20% of the population in developed countries has some disability impacting their ability to walk or use stairs, and research has found disagreement in infrastructure preferences even between broad categories of pedestrians with mobility limitations. AccessMap addresses this by allowing users to customize settings including maximum uphill and downhill incline thresholds, whether to treat raised curbs as barriers, and other infrastructure preferences. As users adjust these settings, the map updates in real-time: sidewalk segments are color-coded based on their accessibility relative to the individual's settings, with inaccessible segments shown as red dashed lines. The system then generates personalized trip plans via a custom routing API that dynamically weights paths based on each user's preferences. At the time of publication, AccessMap served three municipalities in Washington State (Seattle, Bellingham, Mt. Vernon) covering 109,210 sidewalks and 76,561 street crossings, plus 30 public elevators in downtown Seattle used to circumvent extreme inclines.
Key findings
A preliminary user study with 5 participants who self-identified as having mobility limitations compared AccessMap against four alternative mapping approaches for planning an accessible pedestrian route. AccessMap with automatic trip planning scored highest (4.9/5 usefulness, SD=0.22), followed by AccessMap with self-planned routes (4.6/5, SD=0.55), a disconnected assets map similar to Seattle's Accessible Route Planner (3.9/5, SD=0.55), Google Maps (2.2/5, SD=0.84), and a potential barriers point-based map (2.0/5, SD=0.71). The addition of trip planning functionality to AccessMap version 2 drove a 47% year-over-year increase in usage, with over 90% of the growth attributable to mobile users — indicating that pedestrians need on-the-go trip planning rather than desktop pre-planning. AccessMap averages 500-1,000 unique monthly users. The system gathers and transforms data from multiple open sources (municipal open data, OpenStreetMap, USGS) using a scalable ETL workflow, and all data conform to an open schema extending the OpenMapTiles specification. The entire platform — website, data pipelines, and routing engine — is open source.
Relevance
AccessMap addresses a fundamental gap in pedestrian infrastructure: while vehicle routing has been solved at scale by services like Google Maps, pedestrian routing that accounts for mobility disabilities barely exists. A wheelchair user, a person with a walker, and someone with a cardiac condition may all need completely different routes between the same two points — one avoiding steep hills, another requiring curb cuts, a third needing benches for rest stops. The individualized, real-time approach — where users define their own accessibility thresholds rather than being forced into pre-defined categories — reflects the reality that mobility needs exist on a spectrum, not in discrete buckets. For accessibility practitioners and urban planners, AccessMap demonstrates several important principles: the network-first approach (modeling sidewalks as a connected graph rather than isolated points of interest) enables actual trip planning rather than just barrier identification; the open data and open source architecture enables other cities to deploy their own instances; and the dynamic weighting system shows how personalization can be built into routing algorithms. The finding that Google Maps scored only 2.2/5 for accessible pedestrian routing highlights how poorly mainstream mapping services serve people with mobility disabilities — a problem affecting millions of daily trips.
Tags: accessible mapping · pedestrian accessibility · mobility impairment · wheelchair accessibility · wayfinding · open data · personalization · trip planning · built environment accessibility · open source
Standards referenced: OpenMapTiles