A Cyclist’s Encounter with an Indecisive Google Self-Driving Car

A bicyclist recently had a two-minute standoff with a Google self-driving car at a four-way stop in Austin, Texas. So what happened? We explain.

Photo Caption: A Google self-driving car circles the Google campus, sharing the road with bicyclists.

Google’s self-driving cars have been pretty safe up to this point. With more than one million autonomous miles in six years, Google’s self-driving cars have yet to cause any accidents.

The company was just recently involved in its first accident where injuries occurred as the self-driving Lexus SUV was rear-ended in Mountain View, Calif.

There are many obstacles that remain, however, before Google can meet its goal of releasing a self-driving vehicle by 2020. And dealing with bicyclists on the road might be one of them.

MUST-READ: 5 Things You Might Not Know About Google’s Self-Driving Cars

Gregg Tatum, an avid cyclist who typically logs about 6,000 miles per year, recently had an interesting encounter with an indecisive Google self-driving Lexus SUV while riding in Austin, Texas.

He first posted his story on Road Bike Review and retold it to Robotics Trends. Here’s what he says happened:

“Recently, near the end of one of my rides, I was approaching an intersection in my neighborhood that is signed as a four-way stop. As I slowed and prepared to stop, a vehicle entered the same intersection from the street to my left. The white Lexus SUV was unusual as it was adorned with a prominent roof-top camera as well as a variety of other sensors attached to various parts of the car. Closer inspection revealed that it was labeled as a ‘Google Self-driving Car.’

“The car arrived at the stop line a fraction of a second before I did and had the legal right-of-way. Not wanting to unclip my shoes from the pedals, I performed a maneuver called a track-stand in which a rider comes to a stop and then simply balances the bike without putting a foot down.  As I waited for the car to proceed, I noticed that there were two occupants inside that appeared to be observers/testers.

MUST-READ: 18 Pros and Cons of Google’s Restructuring for Self-Driving Cars

“The car remained motionless for several seconds and I continued to maintain my balance without moving. As the car finally inched forward, I was forced to rock the handlebars to hold my position. Apparently, this motion was detected by one of the sensors and the car stopped abruptly. I held the bike in balance and waited for another several seconds and the cycle repeated itself … the car inched forward, I shifted my weight and re-positioned the bars and the car stopped. We did this little dance three separate times and the car was not even halfway through the intersection.  I noticed at one point that the testers were laughing and apparently typing code into a terminal to ‘teach’ the vehicle how to deal with the situation.

“The car finally rolled on and I proceeded on my ride … it was an interesting experience and I noticed that I actually felt safer dealing with a robotically-operated vehicle than one with a human driver.”

Now, Google’s self-driving cars do have patented technology for cyclist hand detection.

The car’s sensors will notice a cyclist among other objects and vehicles on the road. It then measures the distance between the cyclist’s hand and head to decide whether a cyclist is turning or stopping. The algorithm will also look at the angle at which the cyclist’s elbow is bending, and the size and shape of the cyclist’s hands, arms and head.

Google’s newer self-driving prototype and the Lexus SUVs have the same software and sensors. In the video below, scroll ahead to the 1:10 mark, you’ll see a Google self-driving car detect a cyclist and predict that he is about to switch lanes. “The car knows to continue yielding to the cyclist passing by, even when he changes his mind, multiple times,” the video says.

At his TED Talk in March 2015, Chris Urmson, directing of self-driving cars at Google [x], showed a couple examples of how it deals with bicyclists. At 12:15, a bicyclist doesn’t adhere to a red light as the Google car’s light turns green. The car, however, anticipates that the bicyclist is going to cross the intersection and doesn’t move.

 

So, what happened in Tatum’s situation. Brad Templeton, a self-driving car consultant who once worked with Google’s team, offers some insight:

“Quite early in the project they programmed special behaviour at 4 way stops. At most 4 way stops, if you don’t assert your legal right-of-way, others will grab it, even if they came to the intersection later. Likewise, sometimes people will illegally grab your ROW. So you have to change the ‘I will never advance if somebody else is there’ caution to add a little assertiveness. If you don’t, you will sit too long at a busy stop waiting for a clear slot.




About the Author

Steve Crowe · Steve Crowe is managing editor of Robotics Trends. Steve has been writing about technology since 2008. He lives in Belchertown, MA with his wife and daughter.
Contact Steve Crowe: scrowe@ehpub.com  ·  View More by Steve Crowe.




Comments



Log in to leave a Comment



Editors’ Picks

Disney’s Martin Buehler to Deliver RoboBusiness Opening Keynote
Dr. Martin Buehler, Executive R&D Imagineer with Walt Disney Imagineering, will...

How Microsoft’s AI Mastered Ms. Pac-Man
A Microsoft AI system has set the all-time high score for the...

5 Bizarre Uses for 3D Printing
Applications for 3D printing continue to evolve. Here are five uses cases...

Security Robot Fall Shows Limits of AI
Despite all the hype, AI systems get lost all the time, just...