New Video of the Tempe Crash Looks Really Bad for Uber and Its Driver
New Video of the Tempe Crash Looks Really Bad for Uber and Its Driver
There has been enough of speculation virtually how it was possible for a high-tech autonomous vehicle to plow into, and kill, a woman pushing a wheel. Much of the early coverage has been framed by some early comments from the Tempe Police that the woman had apparently jumped out in front end of the car. The Department has at present released dashcam video from the car, and they are really disturbing on a number of levels. The video appears to answer at least some of the questions nosotros posed in our earlier coverage.
The Pedestrian Didn't Sally Of a sudden Out Of Nowhere
Initial statements from the Tempe Police force painted a fairly clear, if ultimately problematic, picture of the pedestrian having suddenly appeared "out of the shadows" — every bit if she had been invisible behind foliage on the median until the last second — and suggested that the resulting collision might have been unavoidable. However, the dashcam video tells a somewhat different story. While the collision itself might take been unavoidable, there should accept been plenty time to at least mitigate the impact past braking or swerving. The woman who was killed is already well out into the roadway as the Uber vehicle approaches. She has already slowly walked beyond the left lane of the two lanes and is in the centre of the right lane even before the vehicle's headlamps are close enough to illuminate her.
This fact lonely should ship warning bells to anyone expecting driverless cars from Uber whatsoever time before long. A woman, pushing a bicycle loaded with her belongings, walks slowly across a four-lane road (two lanes in each direction) and was plowed into without whatever credible slowing, swerving, or other reaction from the auto'southward self-driving systems. You can get a sense for the situation from these frame grabs from a few seconds before the crash. The starting time shows the woman's shoes part mode beyond the lane:
A 2nd frame catch shows her coming into total view:
As some of the readers of our earlier story take noted, the car doesn't appear to accept its high beams on. Whether that is typical for Uber'southward cars, or because there is a car upwards ahead isn't known even so. While non having them on might non have affected the machine'southward own self-driving systems, which rely heavily on lidar, it certainly fabricated it much harder for the safety commuter to run into the pedestrian.
The Safety Driver Was Barely Paying Attention
Companies testing autonomous vehicles made a bargain with regulators to get permission: There will be a safety driver fix to take control when the car'south systems are unable to cope with the state of affairs. The jargon for those situations are disengagements. Unfortunately, there aren't very good regulatory systems for ensuring those drivers are properly trained, instructed in how to act, and monitored for compliance.
Some companies like Waymo have been very serious near training and monitoring drivers. Only based on this incident and the video, Uber is not almost as careful. For much of the preceding interval earlier the crash, the Uber safety driver is conspicuously looking down, and not at the route. Whether she is but staring at her hands, or using some blazon of electronic device, or fiddling with her thumbs is unclear from the video. Likewise, it's hard to tell whether her hands are on the wheel fix to take control, but it certainly doesn't seem similar they are.
The Motorcar'southward Rubber Systems Didn't Intervene
We don't know yet what went on in either the Volvo's advanced rubber systems or Uber'south own autonomous-driving software. But there is no evidence the car (or the driver) attempted to tedious down or have evasive action earlier hitting the adult female. The crash does illustrate some of the limitations of current automatic safety systems, though. For example, the documentation we've institute states that Volvo'south impressive Metropolis Prophylactic system — that can discover both pedestrians and cyclists — is but certified to operate at speeds upward to 30mph. The Uber vehicle was traveling at about 40mph. More traditional vehicle collision avoidance systems might still have been operational at that speed, but they are designed primarily to detect and avoid moving vehicles, not stationary people or objects.
Autonomous vehicle manufacture maven, publisher of Smart Driving Cars, and Princeton professor Alain Kornhauser put the event fairly succinctly for me: "While a human may not take been able to avoid this crash, a well-designed well-working collision abstention system should take at least begun to apply the brakes." Kornhauser went on to make the of import point that if the industry is going to acquire from tragedies like this, information technology's imperative that all the vehicle'south telematics exist made public — including lidar, radar, and whatever other cameras or sensors.
Uber's Self-Driving Tech, Policies Appear to Accept Failed
Overall, the video appears to be fairly damning bear witness that Uber's cars and drivers are not ready for the scale of testing on public roads the company has undertaken. Fortunately, Uber has pulled its test vehicles off the road in response to the crash, but the incident raises a lot of new questions most the company's commitment to safety and the maturity of its self-driving effort. For example, inward-facing dash cams are used at least in part to monitor condom driver performance. A more-responsible company would take fired whatever driver who exhibited the lack of attending we see in this video. Uber doesn't seem to take been concerned.
The Case For Intelligent Regulation Is Even Clearer
It's been articulate for a while that the regulation around cocky-driving vehicle programs has been haphazard, and doesn't always lucifer the risks and rewards. It'due south oft driven by a desire to attract exciting, high-tech research labs, or simply for cities to be viewed as leading border. In particular, electric current regulations take general requirements for how those testing programs should work, only petty or no detail on specific policies around condom driver training, monitoring, and auditing.
Hopefully, the argent lining of a tragedy like this crash will be momentum to improve the regulations almost how cocky-driving technology volition be tested and deployed. At a minimum, more controls over how rubber drivers are trained and expected to carry are needed. Beyond that, there'south no reason that companies shouldn't have to bear witness that their cars tin handle a certain level of complexity on their own before they are allowed on the roads — with or without a human safety driver. That is non unlike the fashion we already exam vehicles for crash prophylactic earlier allowing them on the roads.
Source: https://www.extremetech.com/extreme/266158-video-tempe-crash-public-looks-really-bad-uber-driver
Posted by: coffielddiagestan.blogspot.com
0 Response to "New Video of the Tempe Crash Looks Really Bad for Uber and Its Driver"
Post a Comment