By tapping into the global road network visible across millions of satellite images, and having learned how to extract relevant road features from known datasets, the RoadTagger model is able to enrich poorly tagged digital maps even for parts of the world overlooked by large digital mapping companies.
Indeed, creating detailed maps is an expensive and time-consuming process: Think of the Google car driving around with cameras, combining the collected video data with other data to create up-to-date maps. This is what may enable a GPS system to make the difference between diverging or merging lanes, helping the driver take the right direction. Similarly, incorporating information about parking spots can help drivers plan ahead, while mapping bicycle lanes can help cyclists negotiate busy city streets. Providing updated information on road conditions can also improve planning for disaster relief.
Now, analyzing satellite imagery, RoadTagger uses a combination of neural network architectures to automatically predict the number of lanes and road types (residential or highway) even when roads can be occluded by trees or buildings.
“Most updated digital maps are from places that big companies care the most about. If you’re in places they don’t care about much, you’re at a disadvantage with respect to the quality of map,” says co-author Sam Madden, a professor in the Department of Electrical Engineering and Computer Science (EECS) and a researcher in the Computer Science and Artificial Intelligence Laboratory (CSAIL). “Our goal is to automate the process of generating high-quality digital maps, so they can be available in any country.”
In testing RoadTagger on occluded roads from digital maps of 20 U.S. cities, the model counted lane numbers with 77 percent accuracy and inferred road types with 93 percent accuracy. The researchers are also planning to enable RoadTagger to predict other features, such as parking spots and bike lanes.