Top Ad unit 728 × 90

 

Google's Self Driving Cars Surpass 700,000 Miles of Learning



Can a car be taught to navigate through busy city streets on its own?  That's Google's next quest in its journey to develop a self driving car.  Learning to drive on various city streets involves much more than knowing the traffic system, street signs and where to turn.

The hardest part of driving is moving obstacles; pedestrians and other cars that have minds of their own.  So how can a car be pre-programmed to deal with unknown future events?

According to the latest post on Google blog site, "As it turns out, what looks chaotic and random on a city street to the human eye is actually fairly predictable to a computer. As we’ve encountered thousands of different situations, we’ve built software models of what to expect, from the likely (a car stopping at a red light) to the unlikely (blowing through it)."

Using just one example of fire truck reporting to an incident and coming from behind, it can lead to hundreds of different scenarios.  As a driver in that situation, you will have to evaluate so many moving objects and road setting in order to move out of the way safely.

It seems to me that cars will never be able to drive on its own fully on conventional streets, at least not in foreseeable future.  Perhaps some parts of the country's road network can be upgraded with markers and new conventions that can aid autonomous vehicles.

Perhaps, self-driving cars may have easier way to navigate if they are off the ground.  I think I have seen that in a movie scene, The Fifth Element.

Do you believe one day you will be sitting in a car that can take you to your destination on its own?

Google's Self Driving Cars Surpass 700,000 Miles of Learning Reviewed by Brandon on 12:12 PM Rating: 5

No comments:

All Rights Reserved by Commuter Online © 2014 - 2015

Contact Form

Name

Email *

Message *

Powered by Blogger.