In 2009 Google began work on technology to create a fully autonomous car. These cars would be able to interact with regular traffic and follow the rules of the road without any human interference whatsoever. By 2012 the cars could operate without a driver, but Google always had a human in the driver’s seat to take over if anything went wrong. In 2015 Google launched the “Firefly” self-driving car, built entirely by its own engineers to host the self-driving software and hardware. This tiny two-seater came without any pedals or steering wheel. The autonomous car division branched off from Google and became Waymo, which is still owned and operated by Google’s parent company. Last year Waymo added self-driving technology to minivans and trucks and began to test the cars on real city streets in real traffic situations.
How it Works
Self-driving cars use a range of sensors including a GPS unit, and a navigational system to figure out where they are, how to get where they need to go, and what is around them. The cars use a technology called deliberative architecture, which allows them to create an internal map of their surroundings, and choose the fastest and easiest route through that map to reach their destination. They can sense obstacles, like pedestrians or road construction, and steer or brake as necessary based on the new input. All of the hardware for these computer systems is stored in the vehicle’s trunk.
Why Google Believes Self Driving is Better
Google claims that self-driving vehicles will make the road safer for everyone. They will reduce the number of drunk or otherwise impaired drivers, eliminate distracted drivers, and will always follow speed limits. The cars will allow passengers to spend travel time more constructively so that productive time is not lost on the commute. The human element will be removed from driving, which could leave the streets filled with safe, predictable cars that are always reliable.
Real Road Testing Already Underway
Waymo has been quietly testing their autonomous minivans on the streets of Arizona for almost a year now. The test cars have been run with a live human being sitting in the driver’s seat so that a human could take over the controls if anything went wrong. The tests have been fairly successful. The only accidents involved human drivers running into the autonomous cars, and that was usually caused by the human not expecting the autonomous vehicle to follow traffic rules.
Waymo plans to begin implementing a driverless taxi service in the Phoenix area this year. It has already given rides to select passengers in the last few months. Plans for fully autonomous trucks and the sale of autonomous technology to existing car companies are already underway. Waymo still has some bugs to work out, like what to do if a passenger tries to interfere with the car’s operation, how will the system compensate for dangerous weather, and how insurance companies will handle autonomous cars.