Donate to Science & Enterprise

S&E on Mastodon

S&E on LinkedIn

S&E on Flipboard

Please share Science & Enterprise

Google Granted Patent for Autonomous Vehicle Technology

Autonomous vehicle in DARPA 2007 Urban Challenge (Carnegie Mellon University)

Autonomous vehicle in DARPA 2007 Urban Challenge (Carnegie Mellon University)

The U.S. Patent and Trademark Office awarded a patent to Google Inc. in Mountain View, California for its technology that enables cars to switch from driver to driver-less mode. Patent number 8,078,349 — entitled “Transitioning a mixed-mode vehicle to autonomous mode” — was awarded on 13 December.

The patent covers the processes and devices for a car equipped with the proper technology to change from human-driven to autonomously driven. The technology described in the patent  requires more than a simple flip-of-the-switch, however, to make this happen.

Transitioning the vehicle to driver-less mode, according to the patent, involves stopping the vehicle in a predefined spot it calls a landing strip, where the vehicle detects a reference indicator. That reference indicator, in the form of a QR (two-dimensional) bar code or radio-frequency tag, enables the vehicle to know its exact position.

A key factor in the technology is the ability of the autonomous vehicle to provide more precise directions and control than human driving or currently available technologies can provide. The patent’s landing strip, for example, is one of those enhancements, in this case over current GPS systems. The text notes …

Using exclusively GPS may not give enough accuracy for autonomous vehicle operation. For example, in some cases GPS may only give accuracy to about 30 feet, so additional accuracy may be beneficial. A 30 foot margin of error may be too great to operate an autonomous vehicle, for instance, on a sidewalk or a manufacturing facility with a number of different work stations.

The reference indicator can also provide additional instructions from a simple physical address to geographic coordinates to a URL. The car can then retrieve further directions from the vehicle’s on-board computer or a remote database.

In October 2010, Google revealed that it drove autonomous — but not unmanned — vehicles some 140,000 miles from its headquarters in Mountain View to locations in California including San Francisco, Santa Monica, and Lake Tahoe. Among the engineers that developed the Google test vehicle was Chris Urmson, the technical team leader of the Carnegie Mellon University team that won the 2007 DARPA Urban Challenge (pictured at top) seeking autonomous vehicles that could navigate city streets. Urmson is among the five names listed on the Google patent.

Read more: Engineers Developing Human Driving Model to Reduce Crashes

Hat tip: ConceivablyTech

*     *     *

Comments are closed.