1. Transport
June 7, 2016

Google has added fake engine noises to its self driving cars, and taught them to honk

By Barbara Speed

The biggest problem facing self-driving cars is non-self-driving cars.

Autonomous vehicles have very sophisticated spatial awareness, safety features, and precise geolocation technology. But all that isn’t much good once a human-driven car starts lurching towards it on the motorway. 

As a result, Google’s designers are giving their self-driving cars a facility to help them engage with their analogue counterparts: horns. In the company’s May self-driving car report, it said it had been training the cars to honk appropriately by playing honking sounds inside the cars in traffic, and asking riders to judge if the use was appropriate.

The report notes that thanks to this training, “our self-driving software is designed to recognize when honking may help alert other drivers to our presence”. This includes responding to different situations with “different types of horns”, depending on the urgency.

However, as Fast Company magazine argues, it seems a shame that more automation won’t necessarily mean quieter streets.

The self driving cars are electric, and so essentially noiseless when they drive, but Google has actually added the noise of a combustion engine so that passerby can hear the car and recognise acceleration and deceleration. As the report notes, “For people with visual impairments, the sound of an approaching vehicle can be critical information”.

This leaves us in a bit of a race to the bottom, though: the existing noise pollution around roads means new vehicles need to be artificially made louder to be heard.

Content from our partners
The key role of heat network integration in creating one of London’s most sustainable buildings
The role of green bonds in financing the urban energy transition
The need to grow London's EV infrastructure at speed and scale

This article is from the CityMetric archive: some formatting and images may not be present.
Topics in this article :
Websites in our network