Monday, May 5, 2014

Google: driverless cars now have better understanding of city driving

Ready for some even more vehicle news? We have a great article today that you need to really read. Stay up to this day with all automobile related information and auto transport information right here.

Google explains its driverless cars.

Google is getting closer to its dream of developing a marketable driverless car. The search company said on Monday it had improved the software in its eye-catching project, to take greater account of city driving.

Google’s robot cars have now driven 700,000 miles around Mountain View in California, where the tech company is headquartered.

The company announced that its latest software includes:

• More comprehensive and detailed maps to include curbs and different types of traffic signals.

• Improvements in detecting objects around the vehicle â€" classifying pedestrians and cyclists separately from moving vehicles because they follow different rules of the road.

• The ability to “read” signs as they appear and not just in specific locations â€" for example, a stop sign unfolding from the side of a school bus.

• Modelling likely situations â€" Google has developed software that measure the probability of what is likely to happen in thousands of everyday driving situations.

In a blogpost, the company said:

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). We still have lots of problems to solve, including teaching the car to drive more streets in Mountain View [Google’s home town] before we tackle another town, but thousands of situations on city streets that would have stumped us two years ago can now be navigated autonomously.

Google has equipped about two dozen Lexus RX450H SUVs with its prototype technology, which include sensors that create 3D maps of the cars' surroundings in real time.

The cars are tested with a driver in the driver's seat, ready to take control if anything goes wrong, and a Google engineer in the passenger seat.

No comments:

Post a Comment