Today I listened to the latest podcast from All Tech Considered, found on NPR. Robert Siegel led his team through a real life test of some of the latest traffic apps to see just how useful they are. The goal was to get from Robert's home to his office at NPR as quickly as possible. The apps used were INRIX, WAZE (which is Google owned), Apple Maps, and Robert's own Volkswagen navigation system. They did the test during rush hour traffic to see if the apps could correctly navigate them through heavy traffic. The apps estimated the total time in a range from 23-30 minutes and then Robert and his colleagues were off.
Testing multiple traffic apps at once proves to be an interesting decision right off the bat, as they could not even agree with how he should leave the driveway. The apps pointed them to turn left at a light where they ended up sitting in traffic through two full signal changes. Additionally Google warned them of a serious accident that turned out to not exist. While Robert admits in the podcast that this is not the most scientific test, he says it is also not the most convincing test either. In the end it takes them a little over 30 minutes to arrive at their office.
The second half of the podcast Robert speaks to Ray Resendes from the Virginia Tech Transportation Institute. They make the same trek Robert made previously with his team and discuss the technology used in tracking commute date. Sensors from the road and from service trucks driving the road provide the information for the apps themselves. They discuss the accuracy of the technology, given that sensors are not accurate enough to differentiate between multiple lanes, and how the tracking will get better when cars are the ones providing data as well.
While traffic apps sound like they could be very useful, it's not something I have ever used myself, or even considered using before listening to this podcast. Based off of the available information it seems like traffic apps have some hills to climb before they will be as accurate as people would want. Will people have a problem with their car constantly broadcasting GPS data? For my view, I don't think I would really mind if it meant I could not have to deal with traffic. But then I live in Milwaukee now, where traffic feels like a breeze even at its worst compared to the Seattle freeways I had to deal with for so long. I wouldn't go out of my way to use a constantly updating traffic map, but if it came with my car I would certainly appreciate it, assuming it can do the job well.
This comment has been removed by the author.
ReplyDeleteGreat post. I didn't know apps that help you avoid traffic were already out. It does makes think they will definitely get better as time passes, as the coming of Google's self-driving cars will definitely foster innovation on how these apps work.
ReplyDeleteGreat blog post, but you didn't provide a link to the podcast. I Googled and listened to it. You can also read about it here: http://www.npr.org/sections/alltechconsidered/2015/05/18/407658702/the-tech-behind-traffic-apps-how-well-do-they-work
ReplyDeleteA traffic app would be great, a matter of which 1 works the best. In the morning when I'm rushing to work and I'm not sure if I should take the freeway, travel times posted on the freeway aren't always accurate, they can change if an accident happens, and by the time I get to where I can read the travel times on the freeway, I might already be stuck in congestion. I don't know how they can pick up on an accident that really isn't there. Congestions and road work are things that a GPS can't pick up. It will be interesting to see if traffic apps will catch on and which 1 will be the most popular.
Check out my blog on the Space Tag app at http://mikemendizabal.blogspot.com