You know the feeling when you’re walking to a location in an unfamiliar place and don’t know which way to turn to get there? Google Maps to the rescue. The tool, which is celebrating its 15th birthday, has just released a host of new features, one of which is designed to help in those situations where a command like “head north” doesn’t cut it.
Live View (which was introduced last year) is getting an upgrade. The Street View overlaid with AI, augmented reality, and your phone’s camera will get a boost that shows walkers how far away and in which direction a place is.
Crowdedness (also introduced last year) will incorporate new insights from past riders on all-important details such as:
- Temperature: Will tell you if your train or bus is likely to be hot or cold, according to other riders.
- Accessibility: Offers a listing of public transit lines that have staff to assist, accessible entrances and seating, accessible stop buttons, or high-visibility LED.
- Women’s section: Identifies where transit systems have special women’s sections and notes about whether other passengers abide by the rule.
- Security onboard: To help identify if there will be a security guard, cameras, or an available helpline.
The app itself got an upgrade to organize features under special tabs such as Explore, Commute, Saved, Contribute, and Updates. The latter adds a feed of trending places based on recommendations from local experts and publishers that curate destinations such as the Infatuation. It will also allow you to contact a business directly if you have any questions.
Of course, much of this data relies on users, whether they’re sharing directly with Google Maps or not. And it can get it wrong, as anyone who’s seen the fake traffic jam an artist created by wheeling a cart of 99 phones into a street. However, Google Maps is constantly being updated with new information thanks to AI, according to its senior vice president Jen Fitzpatrick.
For instance, Fitzpatrick writes in a blog post, “We worked with our data operations team to manually trace common building outlines, then trained our machine learning models to recognize building edges and shapes. Thanks to this technique, we’ve mapped as many buildings in the last year as we did in the previous 10.”
Fitzpatrick also says machine learning has been useful in recognizing handwritten building numbers, especially in areas that don’t have formal street signs and house numbers. “In Lagos, Nigeria, alone, machine learning has helped us add 20,000 street names, 50,000 addresses, and 100,000 new businesses—lighting up the map with local places and businesses where there once was little detailed information,” she writes.