Uber

[bannedfromHandV]

2 hours ago, Stevo985 said:

Still not a valid comparison, imo. 

I don’t think there is an equivalent in the retail world. 

Maybe someone like Amazon. 

It was a general comparison in terms of turning a blind eye to the ethics in favour of saving a few £

Not specific to this data protection issue.

[bickster]

Following on from @HanoiVillan Here's part 13...

Quote

Can Uber Ever Deliver? Part Thirteen: Even After 4Q Cost Cuts, Uber Lost $4.5 Billion in 2017
Posted on February 16, 2018 by Yves Smith
By Hubert Horan, who has 40 years of experience in the management and regulation of transportation companies (primarily airlines). Horan has no financial links with any urban car service industry competitors, investors or regulators, or any firms that work on behalf of industry participants

Uber released new financial data this week, showing full year 2017 GAAP operating losses of $4.5 billion, and an operating margin of negative 61%.

Uber’s challenge is to get the business press to downplay or ignore its dismal financial results, and to present a narrative where these results demonstrate that Uber is making major progress. Although Uber has complete financial information, it only released detailed data for three quarters in 2017 and two quarters in 2016. Although much of the missing data can be filled in from earlier releases, Uber knew that reporters on deadline would not do that, and thus key changes over time would not be examined.

Uber’s financial releases have never used consistent definitions of “losses,” often using EBIDTAR or EBIT contributions instead of true GAAP Operating or Net Income. Stories about 2017 results would normally highlight comparisons with 2016, but because Uber has never supplied reporters with 2016 GAAP profit numbers, their stories did not mention that aggregate losses had continued to increase.

Every news report focused on the continued growth in top-line revenue.  Most emphasized the major progress new Uber CEO Dara Khosrowshahi had made towards his goal of a 2019 IPO.

Only one news report, by Eric Newcomer of Bloomberg, highlighted the magnitude of total GAAP losses. Newcomer noted that “[t]here are few historical precedents for the scale of its loss” and explained “Uber prefers to use a different number to refer to its loss: $2.2 billion.” Newcomer recognizes that $4.5 billion GAAP loss was the better indicator of Uber’s current economic performance, but couldn’t explain its superiority to his readers because Uber refused to identify the costs excluded from its preferred number.[1]

The stories that highlighted Uber’s preferred narrative had difficulty linking those claims to the financial numbers that had been released. Emir Efrati led off his story in The Information by saying “Uber is moving toward profitability” even though losses were actually increasing and then emphasized that “[t]he results suggest Uber is getting more efficient as it scales up, allowing it to improve profit margins”[2] without any attempt to document evidence of meaningful scale economies.

The New York Times emphasized that the new numbers “reflect a steady improvement in the company’s financial position, with revenues growing and losses narrowing” but never mentioned the $4.5 billion annual loss (except to say that losses were “big”.)[3] The Financial Times highlighted that “Improvements boost new chief Dara Khosrowshahi as he prepares to take company public” Its lead paragraph reported the smaller EBIDTAR contribution numbers as “losses;” did not mention actual GAAP losses until the end of the story, and made no attempt to explain why it had emphasized the smaller contribution numbers, or why Uber was emphasizing a “profit” measure that excluded $2.3 billion in costs.[4] None of the stories identified the actual magnitude of recent cost savings, or explained them in terms of the overall achievements that would be needed to reach breakeven.

Did Uber’s 4th Quarter cost cuts reflect improved efficiency, or major progress towards profitability?

Uber did cut its costs in the 4th quarter, but a quick review of the reported data raises serious questions about the “Khosrowshahi making major progress towards profitability” narrative. As discussed at length previously, the issue is not whether a startup company might initially incur big losses, but whether its business model demonstrates the powerful scale or network economies that would steadily improve operating margins.[5] The issue is not whether Uber has, and will continue to be able to show in its IPO documents how it can not only reverse $4.5 billion in losses and achieve breakeven, but steadily grow profits for many years to come.

The table below shows quarter-to-quarter changes in revenues and operating expenses, based on the detailed data in Efrati’s article.

Between the 2nd and 3rd quarter of 2016, Uber operating expenses rose 3-4 times faster than Uber’s revenue (the portion of passenger payments that Uber retained).  That is the polar opposite of what one would see if Uber had the kind of scale economies needed to “grow into profitability”. Between the 2nd and 3rd quarter of 2017, revenue growth rates had slowed but operating expenses were still growing just as fast. This represents an improvement over 2016, but unit costs were still not declining.

In the 4th quarter of 2017, Uber suddenly improved unit operating costs, but there is no reason to believe that Uber’s underlying operating efficiency actually improved. In part eleven of this series,[6] based on financial results through the 3rd quarter, I estimated a full year GAAP loss of roughly $5 billion. After years of steady cost increases, Uber froze spending on Operations, Sales and Marketing, Research and Development, and General and Administration. Insurance, an entirely variable cost, continued to increase in line with revenue. 4th quarter costs would have been $210-220 million higher ($800-900 million annualized) if they had continued to grow in line with revenue. These are the kind of across-the-board budget cuts big companies frequently make to stem growing losses.

Perhaps this is an initial step in the right direction, but the budget freeze cannot be said to have seriously improved efficiency unless Uber can demonstrate that it can continue to produce 10-15% quarterly revenue growth on the frozen 3rd quarter 2017 cost base. And then Uber needs to find another $4 billion in cuts..

Twitter just announced profit gains from similar (although larger scale) cost-cutting efforts, but press coverage was largely negative, noting that the cuts threatened the product and market development efforts critical to the long term survival of any rapidly growing technology-oriented company.[7] But unlike Twitter, Uber is a younger, less-mature business that has not yet demonstrated the sustainability of its business model. Twitter’s cost cuts actually produced GAAP profits. By contrast, Uber needs to make another $4 billion in cuts to reach breakeven. Twitter went public in 2013. If Uber wants to go public in 2019, it needs to be able to show that it is continuing to invest in future growth, and that rapid growth will continue to drive margin improvements for many years. Uber’s attempt to cut costs in order to marginally reduce short-term losses directly undermines that narrative.

Is Uber artificially inflating its top-line revenue growth claims?

All previous releases of Uber revenue data were limited to the top-line “Gross passenger payments” (the total money paid by passengers) and “Uber revenue”, the 20-30% of that total retained by Uber. In past analysis, I had assumed that the difference went almost entirely to drivers, but the newly released data shows this assumption is not true, and that Uber may be inflating the top-line revenue number.

In 2017, roughly $3 billion[8] of this revenue was “Refunds, Taxes and Fees” or “Rider Promotions.” Government charges and fares that are refunded should not have been included in the original gross revenue number. The “Rider Promotions” item is more problematic.

If Uber offered discounts, the higher fare (that the passenger did not pay) appears to be included in gross revenue, while the promotional discount is a separate offset.[9] These numbers do not affect bottom line P&L calculations, but inflating the top-line gross revenue number directly supports Uber’s desire to show the strongest possible passenger demand numbers. Uber has steadfastly refused to release any numbers (such as market-specific fare and yield trends) that would meaningfully document whether (or where) its revenue performance might actually be improving.

Uber cut driver compensation by $2.2 billion in 2017

The new data allows one to more precisely calculate recent Uber cuts in driver compensation. In the 2nd and 3rd quarter of 2016, 78% of gross passenger revenue went to “Driver earnings and bonuses.” By the 3rd and 4th quarters of 2017, Uber had unilaterally imposed driver compensation cuts that reduced this to 72%. If the higher 2016 rates had remained in force, drivers would have gotten $2.2 billion more in 2017, Uber’s revenue would have been $2.2 billion lower, and Uber’s GAAP losses would have been much higher than $4.5 billion.

It is perfectly reasonable for reporters to consider year-end financial results in the context of Uber’s drive towards profitability and an IPO, but reporting on Uber has steadfastly ignored drivers (and the vehicles they provide) even though they are far and away the biggest component (85%) of what passengers are paying for.[10] Uber still needs $4.5 billion in profit improvement, but it is unclear that it has much scope to reduce driver compensation much further. Driver turnover has soared.  There are growing anecdotal reports of drivers forced to sleep in their vehicles. Khosrowshahi has aggressively publicized efforts to repair relationships with its drivers. Unless Uber achieves quasi-monopoly control of taxi driver jobs, it is difficult to see how Uber can continue to take larger and larger shares of passenger fares.

Uber’s narrative in 2018

Uber has always been a narrative-driven enterprise. Uber hopes that people overlook the fact that its previous narratives about Uber’s powerful, cutting-edge technological innovations, and its unstoppable march to global industry domination have been exposed as nonsense.

Its current narrative seems to focus on three points; (1) everything bad you might have ever heard about Uber was entirely the fault of Travis Kalanick, and now that he’s gone all those problems have been completely solved; (2) Dara Khosrowshahi is already doing great things to restore the financial promise Uber always had; (3) the core business is still growing strongly and we are marching steadily towards profitability and a 2019 IPO.

As discussed at length in this series, Kalanick personally drove every aspect—good and bad—of Uber’s business model for eight years. Uber could not achieved its meteoric growth without the monomaniacal, hyper-aggressive culture Kalanick created. No one can explain how the bad behavior that culture spawned can be surgically removed from the rest of the business model, or how a less monomaniacal culture can rapidly fuel both strong growth and billions in profit improvement.

While it is too soon to judge his performance, Khosrowshahi has yet to say or do anything to suggest he actually has a plan that will successfully fuel many years of continued growth and the billions in needed profit improvement. Nonetheless, Uber continues to make major efforts to portray him as someone who has already driven major financial gains.

In reality, Khosrowshahi had little to do with the SoftBank investment or Google’s failure to achieve a major victory in its IP theft lawsuit. The SoftBank plan was established by Benchmark. Google simply didn’t have clear-cut evidence that Uber had incorporated stolen IP into its driverless car designs. The Google lawsuit settlement probably could have been achieved months ago, but Uber orchestrated a process that maximized Kalanick’s public embarrassment, and allowed Khosrowshahi to play the role of the “responsible adult” that made all the problems go away. Next year’s P&L will benefit from not having to spend any more money to defend the Google lawsuit, but these gains should not be attributed to Khosrowshahi’s management skills, or improvements in Uber operational efficiency.

Most importantly, the P&L data simply doesn’t support the “major progress towards profitability and IPO” that most press reports highlight. Yes, improvements have certainly been made and will continue to be made at the margin, but sustainable profits will require a combination of much bigger cuts to driver compensation, much bigger cuts to Uber’s own costs, and getting passengers to pay much higher fares. Since no one can define a path to true profitability, Uber continues to emphasize profit targets that exclude several billion of actual costs. Given Uber’s long success in managing its public narrative, the possibility that they somehow muddle through cannot be ruled out, but the available data strongly suggests that profitability mountain is too big, and many of the actions that could help reduce costs would also choke off future growth.

[1] Newcomer, Eric, Uber Quarterly Sales Rose 61% to $2 Billion Amid Heavy Loss, Bloomberg, 13 February 2018

[2] Efrati, Emir, Uber Narrowed Loss in Q4: Full Financial Breakdown, The Information, 13 February 2018 (subscription required). Efrati appeared to be the only reporter that Uber provided with detailed P&L data.

[3]Wakabayashi, Daisuke, As Uber Eyes I.P.O., Its Losses are Slowing. But They’re Still Big. The New York Times, 13 February 2018

[4] Hook, Leslie, Uber pares quarterly losses and lifts revenues, Financial Times, 13 February 2018. There may be legitimate reasons that some of the costs included in Uber’s $4.5 billion GAAP losses do not reflect its ongoing financial situation, but no one has documented the evidence here, and it seems inconceivable that the $2.2 billion negative contribution is the best measure of its situation.

[5] The economic and competitive problems with Uber’s business model are laid out in detail at Horan, Hubert, Will the Growth of Uber Increase Economic Welfare? 44 Transp. L.J., 33-105 (2017)

[6] https://www.nakedcapitalism.com/2017/12/can-uber-ever-deliver-part-eleven-annual-uber-losses-now-approaching-5-billion.html

[7] For example see Wagner, Kurt, Twitter made a profit by cutting costs, not by growing its business, Recode, 8 February 2018

[8] $3 billion is an estimated full year total based on the nine months of actual data released.

[9] Airlines could inflate total revenue by calculating it as “Full Fare” times the number of passengers, and then deducting the discount to the actual fare paid. This is not done because “Full Fares” have nothing to do with actual marketplace prices, and any gross revenue numbers based on them can be easily manipulated.

[10] Taxi service cost structures are documented in section II-B of my Transportation Law Journal article.

16

Naked Capitalism

TL:DR - Uber really are trying to make their figures look good. Included in their revenue figures are the discounts and freebies they dish out. So if someone has a code for £15 discount and the fare comes to £17. Uber put the £17 in the income even though they've only received £2. The there's the giving you the quarters that help their case and leaving out the one's that don't.

The company is essentially a basket case in terms of investment, yet they seemingly intend to launch an IPO next year. You'd be an absolute idiot to invest in Uber.

[bickster]

Wasn't sure whether to put this in here or the things that piss you off topic but this article really gets on my tits

Quote

London-registered minicabs 'working in other cities'

By George GreenwoodBBC News

• 19 February 2018

• Share this with Facebook
•  
• Share this with Twitter
•  
• Share this with Messenger
•  
• Share this with Email
•  
• Share

 

Image copyrightREUTERS

Image captionDrivers for private hire operators like Uber can take jobs far from where they are registered

More than 3,600 Transport for London (TfL)-registered private hire drivers have addresses as far away as Bradford and Cardiff.

Letters released by the Mayor of London's Office show there were 747 TfL-registered minicab drivers in Birmingham and 378 in Bristol.

It means local councils in those areas have no control over drivers' licensing or the number of cabs on their roads.

Sadiq Khan has asked the government for greater licensing powers.

He is calling on the Department for Transport (DfT) for powers to cut the number of licences issued in London and for minicab journeys to begin or end where the driver is registered.

A TfL map included in the letters, and released after a BBC freedom of information request, shows how far these drivers are from the capital.

The DfT said it was monitoring concerns about private hire licensing.

---

Analysis: BBC London Transport Correspondent Tom Edwards

It will come as no surprise to many that minicab regulations are again being shown up as unfit for purpose.

There is meant to be something called a "triple licensing lock" where a job from a passenger can only be accepted if the vehicle, the driver and the booking are under the jurisdiction of the same authority - in this case TfL.

But the lock doesn't work anymore, because minicab apps like Uber means drivers can quite legitimately live and work anywhere in the UK even though they are registered under TfL. You can use the Uber app anywhere to find a car: in the past you'd ring a local number or go to a local shop.

We now have a situation where local authorities across the country, as far north as Blackburn, Bradford and York, have no jurisdiction over TfL minicabs in their own towns.

TfL has even sent officials to places such as Brighton because there are so many TfL drivers there.

The mayor wants a change in the law so the journey has to start or end in the area the minicab is licensed. But at the moment, technology has again outpaced regulation.

---

Image copyrightTFL

Image captionLondon-registered drivers are living far from the capital

Steve Wright, chairman of the Licensed Private Hire Car Association, criticised the mayor's proposed ban on cross-border hiring, saying it would make the current situation worse.

"It would be overly prohibitive, and wouldn't solve the problem of app users showing themselves available in areas far from where they are licensed," he said.

"It would also have a catastrophic effect on the environment, as drivers would have to return to the area they were registered to start new jobs, leading to them driving many dead miles."

A spokesperson for Uber said it would be introducing changes that mean in future drivers will only be able to use the app within the region in which they are licensed.

A mayor of London spokesperson said: "It's time the government gave Sadiq the power to cap these numbers and also urgently end the ridiculous practice of drivers being licensed in one area with the sole intention of working elsewhere in the country."

The DfT said: "The licensing of taxis and private hire vehicles is a matter for local authorities. However we are aware of the concerns regarding the changing landscape of the market, and continue to monitor this."

9

Beeb

So much crap talked 

"But the lock doesn't work anymore, because minicab apps like Uber means drivers can quite legitimately live and work anywhere in the UK even though they are registered under TfL. You can use the Uber app anywhere to find a car: in the past you'd ring a local number or go to a local shop."

Drivers have been legitimately allowed to live wherever they want to since the start of time and work anywhere their operator takes work (The operator must hold a licence for the same borough as the driver). This has been the case since the 1976 LGMPA came into force and introduced the concept of Private Hire to the taxi industry

Living in Birmingham and holding a TfL licence does not mean that the driver is working in Birmingham, he could easily be working in London, this article more than implies that this isn't happening.

 

utter bullshit article tbh,

[bickster]

 

Oops, your personal details are safe with them. Thought they were supposed to have sorted this out.

Roll on 25th May. When the GDPR comes into effect they'll be getting 10/20 million euro fines every week

[bickster]

little bit of snow and ah sod it, lets make some money

[limpid]

32 minutes ago, bickster said:

little bit of snow and ah sod it, lets make some money

Local taxis around here just stopped working. They lasted about two hours after the buses stopped running. I wish we'd had the choice to get gouged.

[bickster]

1 minute ago, limpid said:

Local taxis around here just stopped working. They lasted about two hours after the buses stopped running. I wish we'd had the choice to get gouged.

I guess that's a semi-rural issue. Our driver numbers reduce for sure in snow and ice but we've never actually just stopped.

Most drivers go home because there's too many idiots that don't know how to drive in the conditions

[sne]

Quote

SAN FRANCISCO — A woman in Tempe, Ariz., died after being hit by a self-driving car operated by Uber, in what appears to be the first known death of a pedestrian struck by an autonomous vehicle on public roads.

The Uber vehicle was in autonomous mode with a human safety driver at the wheel when it struck the woman, who was crossing the street outside of a crosswalk, the Tempe police said in a statement. The episode occurred overnight, although the authorities did not specify whether it was late Sunday or early Monday. The woman was not publicly identified.

An Uber spokeswoman said the company was “fully cooperating” with the local authorities. The company said it had suspended testing of its self-driving cars in Tempe, Pittsburgh, San Francisco and Toronto.

 

https://www.nytimes.com/2018/03/19/technology/uber-driverless-fatality.html?action=Click&contentCollection=BreakingNews&contentID=66696641&pgtype=Homepage

[Dr_Pangloss]

Lawsuit time.

[bickster]

54 minutes ago, Dr_Pangloss said:

Lawsuit time.

big queue to be at the back of

Just wait until May 25th and the GDPR introduction, my money's on Uber to be the first company prosecuted

[ThunderPower_14]

Looking forward to seeing whether the fault was with the car or the woman.

[sne]

49 minutes ago, ThunderPower_14 said:

Looking forward to seeing whether the fault was with the car or the woman.

It was a Volvo so the driver (who wasn't there) would have been perfectly safe in the crash.

Silver lining...

[bickster]

6 hours ago, sne said:

It was a Volvo so the driver (who wasn't there) would have been perfectly safe in the crash.

Silver lining...

That's the thing though, there is a human in the cars they are testing, presumably, to override the system should anything go wrong. Failsafe fails in other words

[bickster]

7 hours ago, ThunderPower_14 said:

Looking forward to seeing whether the fault was with the car or the woman.

US law on these matters is somewhat different to ours for example, so any conclusion that is to be drawn wouldn't be applicable in similar circumstances elsewhere

[sne]

Apparently is was a homeless woman, carrying loads of stuff on her bicycle who just without notice walked out into the street.

Quote

Pushing a bicycle laden with plastic shopping bags, a woman abruptly walked from a center median into a lane of traffic and was struck by a self-driving Uber operating in autonomous mode.

“The driver said it was like a flash, the person walked out in front of them,” said Sylvia Moir, police chief in Tempe, Ariz., the location for the first pedestrian fatality involving a self-driving car. “His first alert to the collision was the sound of the collision.”

Traveling at 38 mph in a 35 mph zone on Sunday night, the Uber self-driving car made no attempt to brake, according to the Police Department’s preliminary investigation.

Elaine Herzberg, 49, was unconscious at the scene and later died of her injuries at a local hospital. At a news conference, Tempe police said it appears that she may have been homeless.

The self-driving Volvo SUV was outfitted with at least two video cameras, one facing forward toward the street, the other focused inside the car on the driver, Moir said in an interview.

From viewing the videos, “it’s very clear it would have been difficult to avoid this collision in any kind of mode (autonomous or human-driven) based on how she came from the shadows right into the roadway,” Moir said. The police have not released the videos.

The incident happened within perhaps 100 yards of a crosswalk, Moir said. “It is dangerous to cross roadways in the evening hour when well-illuminated, managed crosswalks are available,” she said.

The Tempe police will collaborate with investigators from the National Transportation Safety Board and the National Highway Traffic Safety Administration in probing the accident.

While hundreds of autonomous cars operate in Arizona, Moir said she was aware of only one other accident, which occurred a year ago. It also involved an Uber in self-driving mode, which was flipped onto its side. But authorities determined that the other car involved was at fault for failing to yield and cited its driver for a moving violation.

“I suspect preliminarily it appears that the Uber would likely not be at fault in this accident, either,” Moir said.

However, if Uber is found responsible, that could open a legal quagmire.

“I won’t rule out the potential to file charges against the (backup driver) in the Uber vehicle,” Moir said.

But if the robot car itself were found at fault? “This is really new ground we’re venturing into,” she said.

https://www.sfchronicle.com/business/article/Exclusive-Tempe-police-chief-says-early-probe-12765481.php#photo-15257361

[bickster]

oops self driving car exceeding the speed limit

[bickster]

Dashcam footage of fatal accident in BBC report here (It stops before the impact)

I think the police statement that was originally issued gave a different impression of the accident

An autonomously driven car should have at least braked

The human "monitor" also clearly wasn't paying any attention to the road

[Davkaus]

On 3/20/2018 at 16:49, bickster said:

That's the thing though, there is a human in the cars they are testing, presumably, to override the system should anything go wrong. Failsafe fails in other words

This is bound to happen, really. A human who's job is to do nothing all day, just in case a very safe system fails isn't an effective failsafe. The more safe a system is, the less often it fails, the more bored and inattentive the person is going to be.

Bit like a keeper trying to stay awake for that one shot in the 90th minute when he's had nothing to do all match.

In this specific example though, looking at the video, I'm not sure a human driver had much chance of stopping in time. Should have certainly at least tried though. There's definitely improvements to be made on this. It was dark but I'm surprised they're not using some sort of night vision?!

Still, even if we chalk this entirely up to a failure of the automated car (and I'm not sure we should), that's one fatality, in how many miles? I'll still take the self driving car over a human driver when it comes to safety.

Edited March 22, 2018 by Davkaus

[bickster]

1 hour ago, Davkaus said:

In this specific example though, looking at the video, I'm not sure a human driver had much chance of stopping in time. Should have certainly at least tried though. There's definitely improvements to be made on this. It was dark but I'm surprised they're not using some sort of night vision?!

Still, even if we chalk this entirely up to a failure of the automated car (and I'm not sure we should), that's one fatality, in how many miles? I'll still take the self driving car over a human driver when it comes to safety.

That car doesn't attempt to stop or swerve, even though there is clearly an impediment to its progress.

Its not a case of even it, its a catastrophic failure

This article in the Wall St Journal is much better at explaining why.

Quote

How the Uber Self-Driving Car Could Have Missed Seeing the Pedestrian

The vehicle should have automatically braked before it hit the pedestrian, a robotics-perception expert said

By 

Tim Higgins

Updated March 22, 2018 12:54 p.m. ET

64 COMMENTS

 

TEMPE, Ariz.—The roads north of Arizona State University are in many ways ideal for testing self-driving cars, with wide, clearly marked lanes and minimal traffic late at night, when the vehicle’s laser sensors work best.

The optimal conditions make it especially troubling that an Uber Technologies Inc. self-driving car plowed straight into and killed a pedestrian walking across a street here at night, without appearing to brake or veer, according to a video from the vehicle released by police Wednesday.

The incident raises questions about whether the sensors that serve as a self-driving vehicle’s eyes are ready for the complexities of city life.

Several autonomous-vehicle experts who reviewed the 21-second video expressed surprise that the car never seemed to detect the 49-year-old woman as she pushed a bicycle across multiple lanes of traffic. They say it is clear the system failed, though some pointed to hazards that can confuse a robot’s brain and eyes, such as shrubs that encroach on the roadway or the woman’s bike with bags on it.

Uber called the video “disturbing and heartbreaking” and said it has suspended testing while cooperating with investigators.

Some experts said the accident might have been more understandable had a person been driving the car. The pedestrian, Elaine Herzberg, was crossing the street in the dark a few hundred feet from a crosswalk, and she seems to come out of nowhere in the video.

But an autonomous car is designed to “see” objects in the dark from hundreds of feet away. The promise of autonomous-car technology is that robot eyes are supposed to detect things humans can’t, and anticipate and react faster to reduce accidents like this one.

“She absolutely should have been detected by their system,” Missy Cummings, a professor of mechanical engineering and material science at Duke University, said Wednesday by email.

Making matters worse, the human safety operator in the car appeared to be distracted, according to the video, and was looking down for five seconds just before impact.

An analysis of the accident’s circumstances points to possible shortcomings of the vision technology that dozens of auto makers and tech giants have hailed as a fundamental part of bringing driverless cars to the road.

The collision occurred as the car was headed northbound on North Mill Avenue, which rapidly widens from two lanes to five as it approaches a vast intersection. Police on Tuesday wouldn’t say precisely where the incident occurred, but the video suggests it happened as the avenue takes a slight curve and then opens into four lanes.

The video confirms the car is in the right lane, possibly giving it a clear view of the other lanes where the pedestrian would have been coming from.

A median with shrubs lines the left side of the road, while the far right lane abuts a bike path and a wide sidewalk lined with some shrubs. On Monday at around 10 p.m., the same time as the crash the night before, the area was quiet with little traffic, though several pedestrians were walking around, including some pushing bicycles with bags.

While police said the speed limit on the stretch where the crash occurred is 35 miles an hour, a nearby sign states 45 mph. The Uber vehicle is believed to have been traveling at about 40, the police said.

Most companies developing autonomous cars, including Uber, use cameras and laser and radar sensors to gather data for the onboard artificial intelligence system to translate. That information then is used to predict what might happen and how the car should respond.

Because it was night, Uber’s system was likely relying heavily on laser sensors, called lidar, which emit beams that bounce off objects to paint a picture of the world, outside experts say. The bulky box of sensors, mounted on the car’s roof, can see in 360 degrees and detect objects hundreds of feet away.

But lidar sometimes struggles to create a complete picture, especially if only a few laser points are bouncing off an object. Snowflakes, for example, can appear to be menacing solid obstacles. It works best at night, when there is no sun interference.

Some driverless-car experts say Uber’s system could have been confused by the shrub alongside the roadway and may not have discerned the difference between shrub branches and a pedestrian. Or the system might not have recognized a woman with a bike and bags and calculated the probability of her walking into its lane.

Others reviewing the video and crash scene say the technology failed. “There’s no question the Uber vehicle should have automatically braked before it hit her,” said Todd Humphreys, an associate professor who specializes in robotic perception at the University of Texas at Austin. He said sensors on the vehicle should have had a clear view of the pedestrian. He estimated the car had twice the amount of distance needed to brake.

Raj Rajkumar, a Carnegie Mellon University professor and founder of an autonomous-vehicle software company that he later sold, suggested the laser sensor may have had a blind spot around the vehicle because of its position mounted on the roof. Still, he said, he would have expected the software to have reacted.

“Legally speaking, the pedestrian may be at fault,” Mr. Rajkumar said. “But mature, reliable self-driving vehicle technology would have done better by slowing down or changing lanes, and this major incident would have been prevented.”

 

[lapal_fan]

The whole technology is in its infancy.

How many people died trying to put planes, helicopters or rockets up in the air? And we still use 2 of those as everyday transport now.

Unfortunately, these incidents were always going to happen.

It's a matter of time until we see driverless cars and I can't wait.