UnLoc uses invisible landmarks to get you around indoors
GPS is awesome to get around with when you’re outside. The minute you get under a roof, though, it’s absolutely worthless. Researchers at Duke University now have an indoor alternative.
UnLoc, short for Unassisted Indoor Location, uses machine learning and “invisible landmarks” to map a building and get a sense of locations.
Here’s an example. Say there’s a dead spot in a building where you lose all signal. To the app, that would be a “landmark” and that allows it to fix certain locations. Similarly, your phone’s various sensors can pick up the vibration of an elevator moving up and down. Also helping? Most phones now come with a compass.
It’s not yet a perfect system; essentially, you’re orienteering your way through the great indoors as the app guides you from landmark to landmark. But as it learns more landmarks, and as it develops a better map of the location, it becomes easier to use, and quickly.
This is a useful idea because mapping huge public spaces is a lot more difficult than mapping roads and wilderness. Most modern maps use satellite photographs, so unless you want to rip the roof off the mall, that’s not going to be useful. This could be a good alternative, and help build an indoor mapping database.
Now, will it help us find a bathroom?