News (Home) | Righters' Blog | Hollywood Halfwits | Our Store | New User Intro | Link to us | Support Us

RightNation.US: Autonomous Uber car strikes pedestrian; woman dies - RightNation.US

Jump to content

Autonomous Uber car strikes pedestrian; woman dies

On Sunday March 18, 2018, an Uber car operating in fully-autonomous mode (with a safety driver) struck a pedestrian in Tempe, AZ. The victim, identified as Elaine Herzberg, later died of her injuries. This is a sad and tragic accident, and my heart goes out to her friends and family.

This is the first known pedestrian fatality involving a fully autonomous vehicle; besides law enforcement officials, both the NHTSA and NTSB will conduct investigations. Tempe police have not ruled out charging the safety driver, who has given a statement to them.

This car was designed to be autonomous (I.E.: driverless), which is not the same thing as the crop of driver-assist technologies now available on the retail market. I've written before about drivers trying to leverage those technologies into "autonomous" driving, which is not the intention or design. This car WAS intended to be driverless.

Shortly after the accident, Devin Coldewey penned this editorial which was published at TechCrunch, titled: "Here's how Uber's self-driving cars are supposed to detect pedestrians" (excerpts):


… Uber's vehicles are equipped with several different imaging systems which work both ordinary duty (monitoring nearby cars, signs and lane markings) and extraordinary duty like that just described. No less than four different ones should have picked up the victim in this case.

Top-mounted lidar. The bucket-shaped item on top of these cars is a lidar, or light detection and ranging, system that produces a 3D image of the car's surroundings multiple times per second. Using infrared laser pulses that bounce off objects and return to the sensor, lidar can detect static and moving objects in considerable detail, day or night…

It's an interesting missive with not-too-technical details; of the four systems, LIDAR seems most relevant in this incident. It was dark, and the victim was crossing a significant secondary road at an unlit and uncontrolled point. Although the role and actions of the safety driver will be debated, I'm not certain that any human driver could have avoided this accident.

How the LIDAR apparently failed remains to be seen.

The next day, Tempe Police Chief Sylvia Moir told the San Francisco Chronical:


… "I suspect preliminarily it appears that the Uber would likely not be at fault in this accident, either…"

The Chief was roundly criticized for that comment, and it has since been walked back, with the Department saying that it does not determine fault in vehicular accidents (sorry, no link). It's probably just a coincidence that the Tempe police have since released footage video of the accident recorded by the vehicle:


The Tempe Police Department has released the first footage of this week's fatal crash involving a self-driving Uber. Two angles of the crash — one facing out at the road, and one facing in at the Uber safety driver — were compiled into a 22-second video that was released on the Tempe Police's Twitter account Wednesday night. In both angles, the footage stops just before the car strikes and kills pedestrian 49-year-old Elaine Herzberg…

There's a link to the video at the end of that article, and I agree that it's "disturbing and graphic" (discretion advised). However, I think it also demonstrates the nearly-impossible context of the accident. I watched the first part several times; and beginning when the car left the lighted part of the road, I counted: "ONE one-thousand, TWO one-thousand"… the accident occurred before I finished.

It looks like the safety driver (and any of us) would have had less than two-seconds to react. At 38-mph, the car was moving at ~55.7 feet per second, going from a lighted to unlit section of roadway. The victim, apparently wearing dark clothing, only becomes visible after "the point of no return". As a driver, I know that I would be held liable (and rightfully so) but I'm reminded of the Kobayashi Maru, the "no win" scenario.

I don't know what the various investigations will reveal, and it will likely be months before the results are released. Yet, my gut feeling is that even a highly attentive drive would have had great difficulty avoiding this accident.

0 Comments On This Entry