Google's Self-Driving Cars Are Self-Honking
Google's last self-driving car monthly report Wednesday shows some exciting update---the project's team is training the cars' AI prototype how and when to honk on its own.
The report shows that engineers who are part of the project are programming its honking algorithm to understand the appropriate time that it needs to honk and also use different honks and beeps for diverse situations.
These circumstances could be diverse---one car backs out of a blind driveway and another goes down the wrong way on a one-way street. In such situations, the car first gives a small honk within the vehicle, with the engineers recording if it needs to make a noise. It would then give some feedback to the AI. Once they feel that it is ready, they will permit the AI to honk on its own.
"Our goal is to teach our cars to honk like a patient, seasoned driver," Google's self-driving car team wrote in the report. "As we become more experienced honkers, we hope our cars will also be able to predict how other drivers respond to a beep in different situations."
Even as the autonomous car gives two short honks to the next car that is slowly reversing toward it, the vehicle can beep louder honks in other serious times.
Moreover, Google's prototype cars have also got a 'hum' sound. They can mimic the sounds of traditional cars that offer pedestrians and cyclists the option to hear them. The hum can increase when the car picks up speed and decrease when it slows down.
A number of tones to create the basic sound, including those of" ambient art sculptures, other vehicles and consumer electronics products and even Orca noises" have been experimented with, even though they have not hit upon the "perfect" sound.
