Google is trying to teach its self-driving cars to cut corners

Oct. 4, 2015

Google’s self-driving cars are too skillful—or at least that’s my conclusion after reading “Google Tries to Make Its Cars Drive More Like Humans” by Alistair Barr and Mike Ramsey in the Wall Street Journal. They quote Chris Urmson, who leads Google’s effort to develop driverless cars, as saying the cars are “…a little more cautious than they need to be. We are trying to make them drive more humanistically.”

One problem is that Google’s cars tap the brakes frequently. Since 2009, Barr and Ramsey report, the cars have been rear-ended 12 times. That exceeds national averages, they add, but national statistics may exclude minor accidents similar to those that Google reports.

Nevertheless, Google is training its cars to “cut corners” and better signal the cars’ intentions to other drivers. One innovation is a “creep” function that the cars can invoke after stopping at an intersection.

That may help. As Barr and Ramsey write, “Some drivers around Google’s headquarters in Mountain View, CA, say they find the cars annoying. They inch forward at T intersections, pausing regularly to get their bearings. They’re extra careful when turning left because their software calculates a minimum safe turning distance from oncoming traffic. Humans have a more instinctive, and higher risk, assessment.”

Read the complete Journal article here.

About the Author

Rick Nelson | Contributing Editor

Rick is currently Contributing Technical Editor. He was Executive Editor for EE in 2011-2018. Previously he served on several publications, including EDN and Vision Systems Design, and has received awards for signed editorials from the American Society of Business Publication Editors. He began as a design engineer at General Electric and Litton Industries and earned a BSEE degree from Penn State.

Sponsored Recommendations

Comments

To join the conversation, and become an exclusive member of Electronic Design, create an account today!