New algorithm lets people with disabilities customize routes in real time instead of choosing from preset options

Researchers developed a route-planning system that learns what matters to each user—steep hills, curb cuts, crowded sidewalks—and adapts suggestions based on feedback, rather than forcing them to pick from a fixed list of pre-computed options. This means someone in a wheelchair, someone using a cane, and someone with a stroller can each get genuinely different routes that actually match their needs, not a one-size-fits-all approximation.

Current route planners (Google Maps, Apple Maps) optimize for speed or distance for everyone equally. They don't know that one person can't use stairs, another needs to avoid steep grades, a third wants well-lit paths at night. This algorithm treats accessibility as a real, learnable preference rather than a constraint to work around after the fact. What becomes possible is a routing system that gets smarter about you the more you use it—and more importantly, that treats disabled users as the design center, not an afterthought.