[infurl]

https://analyticsindiamag.com/google-maps-deepmind-machine-learning-research/

Researchers at Alphabet-owned DeepMind have partnered with the Google Maps team to improve the accuracy of the real-time ETAs by up to 50% in places like Berlin, Jakarta, São Paulo, Sydney, Tokyo, and Washington D.C. They are doing so by using advanced machine learning techniques, including Graph Neural Networks.

In some countries punctuality doesn't matter very much. It's ok as long as you turn up on the day. In other countries it's considered impolite to be five minutes late for a meeting and if you're going to be fifteen minutes late you might as well cancel. The ability to predict travel times accurately can have a significant economic and social impact, so these kinds of improvements to Google Maps are pretty important for anyone living in the real world.

[frankinstien]

If you haven't used google maps for anything else but just find an urban route then you really don't get how inaccurate google maps are. I was hiking in the Owen Gorge area in California on the Labor Day weekend. I marked the camping site I landed on with google maps, which is a BLM area remote and rugged, so I wouldn't get lost as I hiked the area. Normally hikers mark the area with colorful flags so they can get their sense of direction back as they site each flag. I felt very confident that with google maps I wouldn't have a problem. I hiked around and everything pretty much looks the same, trees, large boulders, grass-like plants everywhere. As I headed back I kinda felt a little lost so I turned on my phone and google plotted a path back to camp. The path stated I was moving in the opposite direction and I was 1 hour and 21 minutes from camp! I looked up at the sun and knew I was traveling west generally and when I headed back I was heading east. I didn't trust google maps and decided to rely on my sense of direction and continued east. Well, I was right and google was dead wrong!

[yotamarker]

do they have a rest API ?