Enrich digital maps with artificial intelligence |
MIT researchers have created a new digital map enrichment model called RoadTagger that uses satellite imagery to define road features in digital maps that can help improve GPS navigation. Model changed. Artificially explained in an article by the Intelligence Enhancement Association. ,
GPS maps are often created by large companies such as Google that send cars to vehicles equipped with cameras to navigate and capture videos and pictures of regional roads. However, this process is very expensive and expensive. Of these cards the case is taking too long.
Depending on the high cost, certain parts of the world are ignored and only basic GPS data is shown. One way to solve these challenges is to find a machine model for satellite imagery that can be easily accessed and regularly updated to automatically determine path characteristics.
The problem is that street satellite imagery is often bewildering by trees and buildings, making MIT at MIT a new innovation.
The new model uses a hybrid neural network architecture to automatically predict the number of lanes and road types (residential areas or roads) behind obstacles. The model tested in an isolated way can calculate the number of lanes with an accuracy of 77% and May, and the type of stable street (residential street or highway) is accurate by 93%.
Researchers are also planning to allow RoadTagger to predict other features, such as parking lots and bike paths, and hopefully one day they will be used to help people quickly review ongoing road adjustments.
"Most of the updated digital maps are used when big companies take care of them when they're not in the place they care," said Sam Mosten, co-author, associate professor at EECS in Electrical Engineering and Computer Science and researcher at the Computer Science and Artificial Intelligence Laboratory (CSAIL). Then your surroundings do not affect the quality of the map.
He added: "Our goal is to automate the process of creating high-quality digital maps that can be used in any country."