Photo: Google
No less than six years have passed since the Google Self-Driving Car Project started testing on public roads. During this time span, the fleet of vehicles that make up the project drove 1,419,672 miles (2,284,740 kilometers) in autonomous mode.
The number mentioned above pales in comparison to the three million simulated miles (4.828 million kilometers) Google’s cars drive each and every day. The question is, though, what exactly is the meaning of simulated miles? According to the company, a powerful simulator calls upon the 1.4 million miles driven on public roads to drive those miles again in the virtual world in order to make improvements to the software that goes with Google’s autonomous cars.
Case in point: “To make left turns at an intersection more comfortable for our passengers, we modified our software to adjust the angle at which our cars would travel.” Google tests changes like this by rerunning its public road driving history with the new turning pattern to make sure the car turns left better than it did before. The company’s purpose with these millions of simulated miles is to offer a superior driving experience.
One of the biggest drawbacks of the simulator is that they can’t make Google’s cars better in rain or snow. On the plus side, simulators can create new driving scenarios. In a three-car merge situation, the simulator can draw up thousands of variants of the situation, then test if the car drives as intended in each variation of the given scenario.
Cars traveling at different speeds, the angle of merging, and other variables are the name of the game here. At the end of the day, the Google simulator is just a tool made to speed up the development process of the self-driving car.
Driving AI is, without a shadow of a doubt, one of the coolest technologies being developed at the moment. Given the fact that the California Department of Motor Vehicles recently issued the first proposed regulations governing autonomous vehicles, self-driving cars will become reality faster than initially thought. Check the attached PDF file to read more on the Self-Driving Car Project’s January 2016 report.
Case in point: “To make left turns at an intersection more comfortable for our passengers, we modified our software to adjust the angle at which our cars would travel.” Google tests changes like this by rerunning its public road driving history with the new turning pattern to make sure the car turns left better than it did before. The company’s purpose with these millions of simulated miles is to offer a superior driving experience.
One of the biggest drawbacks of the simulator is that they can’t make Google’s cars better in rain or snow. On the plus side, simulators can create new driving scenarios. In a three-car merge situation, the simulator can draw up thousands of variants of the situation, then test if the car drives as intended in each variation of the given scenario.
Cars traveling at different speeds, the angle of merging, and other variables are the name of the game here. At the end of the day, the Google simulator is just a tool made to speed up the development process of the self-driving car.
Driving AI is, without a shadow of a doubt, one of the coolest technologies being developed at the moment. Given the fact that the California Department of Motor Vehicles recently issued the first proposed regulations governing autonomous vehicles, self-driving cars will become reality faster than initially thought. Check the attached PDF file to read more on the Self-Driving Car Project’s January 2016 report.
Google News