The transport company suspended its tests in Toronto and several US cities last March after the death of a pedestrian snapped up by an autonomous car Uber in Arizona.
The investigation showed that Uber’s employee on board was watching a television program prior to the accident and braked too late. In addition, the emergency braking system of the vehicle was not activated.
Raquel Urtasun, chief scientist for Uber in Toronto, says the company has re-evaluated its procedures and has implemented several additional security measures since the accident.
According to the new road test program, a second employee will accompany by precaution the driver, who must be behind the wheel at all times.
Artificial intelligence and maps
Rather than focusing on autonomous driving capabilities, the test program in Toronto is more about creating better road maps using artificial intelligence technologies.
The company’s autonomous vehicles will drive on both highways and streets in Toronto.
To advance, vehicles of this type depend, in addition to the sensors they have, detailed road maps.
But card creation is currently a very expensive and time-consuming process , requiring multiple trips to the same locations to collect data, says Urtasun.
Uber, she adds, is seeking to develop an automated technology to create real-time road maps using artificial intelligence.
As our second lead editor, Zelda Wiggins provides guidance on the stories Nebula Electronics reporters cover. She has been instrumental in making sure the content on the site is clear and accurate for our readers. Zelda prides herself on having clear and concise headlines without resorting to click bait titles. Zelda received a BA and and MA from Fordham University.