Works when the signal drops
Places you have already looked at stay available in car parks, lifts and basements.
Scroll sideways to follow the app from opening it to building up your profile. Every picture here is the real Android build.
Search, filters and browsing all work off the same accessibility data, so a "nearby restaurant" result you can't get into never rises to the top.
The accessibility score isn't a star rating someone typed in. It's computed from the venue's verified features, weighted by how much each one matters to a wheelchair user, and adjusted by what the community has reported recently.
Venue pages also carry a photo gallery, real opening hours with an open/closed state, reviews with per-aspect sub-ratings, and one-tap directions or sharing.
Three transport modes share one routing engine, then diverge on the things that matter: speed, distance limits and how barriers are scored.
Routes at a realistic rolling speed, refuses journeys beyond your configured maximum walk distance, and prefers paths that cross the fewest reported barriers.
Traffic-aware driving times from a live routing API, with a free fallback engine that takes over automatically if the quota runs out, so navigation never just stops.
Turn-by-turn text-to-speech, plus spoken warnings when you approach a crowdsourced barrier on your path.
Smooth position updates, a live speed badge and a single-shot auto-reroute the moment you leave the planned line.
Point the camera down the street and a directional arrow floats over the live view, fusing GPS heading with the compass so it stays pointed at your next turn while you move. Approaching barriers are announced out loud.
Being straight about scope: this is sensor-fusion guidance, not computer vision. The app does not detect obstacles from the camera image, the hazards it warns you about are the ones the community has reported. That boundary is deliberate, and the code is structured so a real detection source can slot in later.
Three contribution types, each designed to take under a minute, because a reporting flow that takes five minutes never gets used from the pavement.
A day out is a chain of places, and one inaccessible link breaks it. The planner treats the whole chain as the unit.
Crowdsourced data only works if contributing is worth it. Points are awarded by the database itself, never by the app, so the ledger can't be gamed from a modified client.
There is no payment gateway anywhere in Move My Way. Rewards are bought with points only, nothing is ever charged to a card or wallet.
The largest module in the app, at 25 files. A text-and-voice assistant that answers from your own data rather than from a model's memory, and keeps answering when the network gives up.
Crowdsourced data lives or dies on whether the evidence is real. Photos attached to a barrier report are checked in two layers before they can be submitted, cheapest first.
It is a quality gate, not a security control. It runs on the phone, so it removes careless mistakes and misuse. A person still reviews anything the check is unsure about.
Places you have already looked at stay available in car parks, lifts and basements.
Large touch targets, high contrast text and a dark mode that follows your phone.
Tell the app your wheelchair type and how far you are happy to travel, and your routes follow suit.
Location is used while you are navigating and nowhere else. No tracking in the background, no ads.
You hear from the app when you earn something or finish a mission. Nothing else.
Every state and federal territory, with local place names, local malls and local transport.