Ready for Robo-Cars? Tesla’s Self-Driving Scramble has Some Experts Worried

The Virginian-Pilot (Norfolk, VA)

August 30, 2019


Tesla is racing to be first to the market with a self-driving car made for the masses, promising to send as soon as this year an over-the-air software update that will turn hundreds of thousands of its vehicles into robo-cars.


But its push to put untested and unregulated features in the hands of its drivers is putting industry executives and regulators on edge.


Once the update arrives, Tesla vehicles will be able to drive themselves in a city the way they can perform highway cruising now, the company said. That means interpreting stop signs and traffic lights, making sharp turns and navigating stop-and-go urban traffic and other obstacles – a far more difficult task than navigating long, relatively straight stretches of highways.


The electric car manufacturer said it will do that without light detection and ranging – or lidar – complex sensors that use laser lights to map the environment, technology that most autonomous vehicle makers consider necessary. Even with lidar, many of those manufacturers have adopted a slow and deliberate approach to self-driving vehicles, with limited testing on public roads.


Tesla shows little sign of such caution. And because autonomous vehicles are largely self-regulated, no one can stop the automaker from moving ahead.


The Washington Post spoke with a dozen transportation officials and executives, including current and former safety regulators, auto industry executives, safety advocacy group leaders and autonomous-vehicle competitors.


In interviews, they expressed worries that Tesla’s plan to unleash robo-cars on the road on an expedited timeline could result in crashes, lawsuits and confusion. Plus, they said, Tesla’s promised “full self-driving” features fall short of industry standards for a true autonomous vehicle because humans will still need to be engaged at all times and ready to intervene in the beginning.


“That concern among the industry writ large is real and valid because what potentially happens is you’re going to see fatalities in the news attributed to Tesla vehicles and the response you’re going to get from certain policymakers – kind of a knee-jerk reaction,” said a former senior official with the National Highway Traffic Safety Administration, which oversees the motor vehicle industry, who spoke on the condition of anonymity. That, in turn, will affect “other manufacturers who were a lot more deliberate, a lot more careful.”


Tesla has said it already has better real-world data than the rest of the industry. The company’s artificial intelligence program is being trained in real time by data collected from every Tesla already on the road. Every touch of the steering wheel helps inform the company’s software of how to react to various scenarios.


Tesla, which launched its first consumer vehicle just over a decade ago, was founded with the goal of bringing electric cars to the masses. It has outpaced most rivals for years, launching cars that have a range of up to 370 miles. Its Autopilot system, which keeps cars within their lanes, performs steering functions and can summon and park cars without the drivers controlling the steering wheel.


Tesla Chief Executive Elon Musk wants to morph that product into his “full self-driving” suite, through a combination of the hardware already in its cars and over-the-air software changes that would add increased capabilities for city driving.


The company has also said that it has a demonstrated track record of safety, registering just one crash for every 2.87 million miles in which drivers had Autopilot engaged in the first three months of the year. That compares with normal cars crashing every 436,000 miles. But Autopilot is intended for use on highways and freeways, relatively uncomplicated roads with long straightaways that have fewer crashes, so it is unclear how comparable those statistics are. Tesla has declined to release more-detailed data.


Tesla cars also would eventually connect to the Tesla Network, equipping them to give rides when their owners aren’t using them, similar to the ride-hailing services of Uber and Lyft.


“By the middle of next year, we’ll have over a million Tesla cars on the road with full self-driving hardware,” with the ability to find the vehicle owners, drive them to their destination and park the vehicle, Musk said at Tesla’s Autonomy Investor Day in April. It will be at “a reliability level that we would consider that no one needs to pay attention … meaning you could go to sleep.”


Meanwhile, competitors are racing to build their own autonomous taxi fleets expected to transport people without drivers within a few years. Companies including Waymo, owned by Google parent Alphabet, as well as Lyft-backed Aptiv and GM Cruise are piloting autonomous vehicles in Arizona, Nevada and California.


Tesla is betting it can win the race with its software updates. Its approach represents a stark departure from the more conservative approaches by many companies testing self-driving cars. For instance, when Uber’s self-driving vehicle hit and killed a pedestrian, the company halted testing of its vehicles for months.


Tesla has raised eyebrows with its statements that autonomous driving can be achieved through a slimmed-down system that sheds all but the most critical equipment. Musk says he wants Tesla’s system to use a combination of cameras and radar sensors that triangulate a field of vision, similar to human eyesight, forgoing lidar.


Tesla executives said at an April conference that the company is using its radar and cameras to understand depth around its cars and real-world road conditions, as well as its Shadow Mode, which allows it to test how self-driving technologies perform without actually activating those features – something the company says lets it train and refine its networks without needing to do the same testing as other companies.


“Lidar is lame,” Musk said. Rivals are “all going to dump lidar. That’s my prediction. Mark my words.”


Faiz Siddiqui; Washington Post


Copyright 2019 Virginian-Pilot Companies LLC. All Rights Reserved.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.

Toyota Announces Recall of Faulty Airbags Again

The Columbus Dispatch (Ohio)

August 30, 2019


Toyota is recalling 191,000 cars in North America and Japan because the air bags may not inflate properly in a crash.


The recall covers certain 2003 through 2008 Corolla compact sedans and 2005 through 2008 Matrix hatchbacks.


Toyota says the front passenger air bags in the cars were installed in previous recalls to replace dangerous Takata air bags.


In high temperatures, the replacement bags might not unfold as designed, increasing the risk of injury in a crash. Toyota wouldn’t say if anyone has been hurt due to the problem.


Copyright 2019. The Columbus Dispatch. All Rights Reserved.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.

Toyota and Suzuki Plan to Collaborate on Development Projects and Autonomous Driving Technology














Click here to read the entire article: Quality News Today.

Safety Tests Delay GM Self-Driving Cars; Company Still Cites Progress on Cruise

USA Today

July 31, 2019

By Jamie L. LaReau


General Motors is delaying the launch of its self-driving cars citing the challenge to safely remove people from behind the wheel.


About 18 months ago, the automaker had promised that its self-driving unit, GM Cruise, would launch a fleet of autonomous vehicles in the form of a ride-sharing service by the end of this year.


But in a blog post Wednesday, GM Cruise CEO Dan Ammann said while the automaker has made strides in hiring key talent and gaining the financial backing needed to continue testing the cars, in order to reach the level of performance and safety validation required to deploy a fully driverless service in San Francisco, we will be significantly increasing our testing and validation miles over the balance of this year, which has the effect of carrying the timing of fully driverless deployment beyond the end of the year.


The additional testing will provide GM Cruise with crucial operational learnings from running a larger scale fleet and ride service, which it currently runs for its employees, said Ammann.


It also means San Francisco residents will see more of the Cruise AVs on the roads being tested, he said. Cruise owns nearly 40% of all of the electric vehicle fast chargers in San Francisco, he said. It is building the largest EV fast charger station in the country there.


GM bought GM Cruise in 2016. Since then, it has been aggressively testing the electric self-driving vehicles in San Francisco and rapidly expanded. It has grown from a staff of 40 people to nearly 1,500, Ammann said in the blog, hiring top engineers.


Cruise also has raised about $7.25 billion from technology investors such as the SoftBank Vision Fund and T. Rowe Price and GM and Honda.


This gives us the deep resources necessary to scale our services in San Francisco and beyond, Ammann wrote.


Still, auto industry experts had expressed doubt this year that GM Cruise could safely put self-driving taxis on the road by the end of this year.


But despite GM idling five of its plants in North America this year and cutting about 14,000 hourly and salaried jobs to save $6 billion by 2020, its commitment to electric and autonomous vehicle technology has been unwavering. GM leaders often refer to GMs desire to be a technology company that builds cars, creating a future with zero crashes, zero emissions and zero congestion.


In February, GM CEO Mary Barra said that GM continues to make rapid progress with the technology when asked if GM would launch a fleet of self-driving cars by year end.


Pressed again for a launch date in April after GM reported earnings, Barra said: I think youll see updates later this year. We are very pleased with where were at on our continued rate of progress. Were doing our testing in one of the most complex environments in the U.S. (San Francisco). We have a very strong position and we have a very strong safety record.


In his blog, Ammann said that GM is uniquely positioned to be a leader in the self-driving space. The company is running neck-and-neck with Waymo, a subsidiary of Google, to be the first to bring the robot car ride sharing to market. The two companies are considered the leaders in a crowded field, and San Francisco is the proving grounds to refine the technology.


But, wrote Ammann, Cruise is the only company with self-driving cars that are manufactured on a large scale automotive assembly line to the same rigorous standards of safety and quality as any other production car.


GM builds the Cruise AV cars at its Orion Assembly Plant in Lake Orion, Michigan. The first Cruise AVs were Chevrolet Bolts, Chevrolets all-electric vehicle, that used two Lidar sensors on the roof. GM Cruise built 130 of the second-generation of its AVs in 2017 and those had a bigger suite of sensors. Then, later that year, it added another 50 of the cars. GM no longer calls them Bolts, but rather the Cruise AV.


This is not a concept carhundreds of the best Honda, GM and Cruise engineers are working together on-site in Warren, Michigan, where they are deep into the vehicle development process, wrote Ammann. This new vehicle completely re-imagines from the ground up what a car can be and we cant wait to share more in the near future.


Copyright 2019 Gannett Company, Inc. All Rights Reserved.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.

The Two-Part Manufacturing Test Under the Buy American Act

Mondaq Business Briefing

August 15, 2019

By Edward (Teddie) Arnold


Congress enacted the Buy American Act (BAA) during the Great Depression, in order to protect American industry from foreign competition on federal procurement contracts. While the BAA is simplistic in its policy goal of promoting domestic purchasing, government contractors and subcontractors are often faced with complex and confusing rules for compliance.


The operative language of the BAA provides:


Only unmanufactured articles, materials, and supplies that have been mined or produced in the United States, and only manufactured articles, materials, and supplies that have been manufactured in the United States substantially all from articles, materials, or supplies mined, produced, or manufactured in the United States, shall be acquired for public use unless the head of the department or independent establishment concerned determines their acquisition to be inconsistent with the public interest or their cost to be unreasonable.[1]


The Government flows down BAA requirements to government contractors in the form of standardized clauses contained in the Federal Acquisition Regulation (FAR) and Defense Federal Acquisition Regulation Supplement (DFARS).


The key to understanding the BAA is determining whether the solicited goods or end products are domestic, i.e. were mined, produced, or substantially manufactured in the United States. The FAR defines an end product as those articles, materials, and supplies to be acquired for public use.[2] The analysis of whether an end-product qualifies as domestic is done using a two-part test.[3] First, the end-product must be manufactured in the United States; second, more than 50% of all component parts (determined by cost of the components) must also be manufactured in the United States. If a product meets this two-part test, then it can be considered a domestic end product under the BAA. End products that do not qualify as domestic under this test are treated as foreign.


The DFARS establishes a slightly difference scheme than the FAR with regard to the BAA. While the FAR only considers domestic production, the DFARS considers both domestic production and qualifying country production. The reason for this distinction is that the Department of Defense (DOD) has negotiated its own trade agreements with certain countries.[4]


Determining whether a product was manufactured in the United States can be challenging. Neither the BAA nor the FAR/DFARS defines the term manufacture or manufacturing. However, as established by published decisions of the Government Accountability Office (GAO), the lynchpin of determining whether a good was manufactured domestically is whether there was a substantial change in the physical character of the good, whether there were separate manufacturing stages involved, or whether there was one continuous process. While the GAO has affirmatively excluded activities like packaging or testing from the manufacturing process, the determination of whether a particular article was domestically manufactured must be done on a case-by-case basis.


Questions often arise as to whether end products that are manufactured in the United States solely from foreign material can qualify as domestic end products. Consider a hypothetical scenario in which a prime contractor engages a subcontractor to provide steel plates that the prime contractor will incorporate into an end product being manufactured for DOD. The prime contractor flows the requirements of DFARS 225.252-7001(c) to its subcontractor, requiring the subcontractor to deliver only domestic end products. The subcontractor procures steel from China (a non-qualifying country under the BAA), manufactures the steel in the United States, and provides it to the prime contractor. Is the subcontractor in compliance with the BAA requirements of its subcontract. The short answer is, it depends on the various manufacturing processes employed by the subcontractor.


In cases where an end product is manufactured in the United States solely from material from other than the United States or a qualifying country, GAO has held that the end product can qualify as a domestic end product if there are two manufacturing phasesboth occurring in the United Statesthe first of which transforms the material into a component and the second of which manufactures the component into the end product. [W]here it is alleged that a foreign material has been manufactured into a component domestically and the component in turn manufactured into an end item domestically, [the GAO] ha[s] also looked at where the manufacturing process consists of two distinct phases, the first yielding a component that is distinguishable from the original material and the second yielding an end item that is distinguishable from the component.[5] Thus, as it pertains to the hypothetical scenario involving the manufacture of foreign steel, the question becomes whether the manufacturing process employed by the subcontractor satisfies the two-part test for end products manufactured solely from foreign materialthat the material has to be transformed first to a component, and then to an end product in distinct phases.


Case law in this area can be somewhat instructive, although sometimes less than clear. Fortunately, for contractors attempting to understand whether two manufacturing processes have taken place, GAO has taken a practical approach based on the facts of each case. For example, in Davis Walker Corp., GAO held that foreign galvanized steel was subject to two distinct manufacturing processes, where the first process involved taking foreign steel rods and converting them into bright wire, and the second process involved taking the bright wire and converting it into galvanized steel.[6] Similarly, in Yohar Supply Co., GAO found two manufacturing processes to have taken place when the company stamped out components from rolled sheets of foreign steel, bent these components into shape, and smoothed them.[7]


In another case, however, GAO found that first taking foreign sulfadiazine and converting it into tablet form, and second, putting the tablets into bottles, constituted a single manufacturing process.[8] GAO reached a similar conclusion in a case involving solvent dye. The alleged manufacturing process consisted of first modifying the foreign material through a series of processes into a processed mixture, and second, taking the processed mixture and sifting and selecting the contract compliant material from the processed mixture. The GAO held that the purported second step of sifting and selecting was not a manufacturing process.[9]


Contractors at any tier attempting to ascertain whether their end products are compliant with the BAA should be asking specific questions of their own processes or their subcontractors processes. Having a clear understanding of the source of the material, the contractors supply chain, and the manufacturing processes used to create the end product will greatly assist contractors in determining compliance with their contracts.


Footnotes

  1. 41 U.S.C. 8302(a)(1).
  2. 48 C.F.R. 25.003.
  3. 48 C.F.R. 25.101
  4. 48 C.F.R. 252.225-7001.
  5. City Chemical, LLC, B-296135.2, B-296230.2, 2005 CPD Para. 2005.
  6. B-0184672, 76-2 CPD Para. 192 (1976).
  7. B-225480, 87-1 CPD Para. 152.
  8. To S.F. Durst & Co., Inc., 46 Comp. Gen. 784, 784 (1967). The GAO reached a similar conclusion in City Chemical, LLC, B-296135.2, B-296230.2.
  9. Id.; but see Marbex, Inc., Comp. Gen. Dec. B-225799, 87-1 CPD Para. 468 (sterilization of gloves has been held not to constitute manufacture because there was no material alteration of the product); Blodgett Keypunching Co., Comp. Gen. Dec. B-153751, 76-2 CPD Para. 331 (keypunching information on a magnetic tape was not manufacture).


The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.


Copyright 2019 Mondaq Ltd. All Rights Reserved.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.

ANSI Expands Cannabis-Related Standards and Conformity Assessment Activities to Foster Health, Safety, and Quality

States News Service

August 15, 2019

In support of its mission to further voluntary standards and conformity assessment activities as a means of enhancing public health, safety, welfare, and the environment, the American National Standards Institute (ANSI), and its subsidiary, the ANSI National Accreditation Board (ANAB), are pleased to announce their intent to strengthen, reinforce, and expand their respective accreditation activities in areas relating to cannabis and cannabis-related products.


In keeping with legal guidelines adopted by the ANSI Executive Committee of the Board of Directors on July 24, 2019, ANSIs activities will include accrediting qualifying standards developing organizations (SDOs) and approving SDOs cannabis-related standards that qualify as American National Standards (ANS). Such standards may address, for example, indoor air quality in marijuana grow operations or standards for the remediation of illicit drugs.


ANAB has provided accreditation offerings for cannabis-related products since 2014, and building upon this significant technical experience, it will continue accrediting qualifying certification bodies, testing laboratories, proficiency test providers, and the like.


The legal cannabis market is incredibly complex, and undoubtedly growing, prompting the need for effective voluntary standards and conformity assessment activities to support health, safety, environmental protection, and supply chain quality, said ANSI president and CEO S. Joe Bhatia. ANSI has a responsibility to facilitate the accreditation and standardization solutions required to promote safety in conjunction with this growing industry, while strictly complying with all applicable Federal and State laws.


About ANSI


The American National Standards Institute (ANSI) is a private non-profit organization whose mission is to enhance U.S. global competitiveness and the American quality of life by promoting, facilitating, and safeguarding the integrity of the voluntary standardization and conformity assessment system. Its membership is comprised of businesses, professional societies and trade associations, standards developers, government agencies, and consumer and labor organizations.


The Institute represents and serves the diverse interests of more than 270,000 companies and organizations and 30 million professionals worldwide. ANSI is the official U.S. representative to the International Organization for Standardization (ISO) and, via the U.S. National Committee, the International Electrotechnical Commission (IEC). For more information, visit www.ansi.org.


About ANAB


The ANSI National Accreditation Board (ANAB) is the largest multi-disciplinary ISO/IEC 17011 accreditation body in North America with an accreditation portfolio that includes management systems certification bodies, calibration and testing labs, product certification bodies, personnel credentialing organizations, forensic test and calibration service providers, inspection bodies, police crime units, greenhouse gas validation and verification bodies, reference material producers, and proficiency test providers.


Copyright 2019 States News Service.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.

Manufacturers Won’t Bear All Liability for Driverless Vehicles

Mondaq Business Briefing

August 29, 2019

By Geoffrey Wyatt


The anticipated impact of autonomous vehicles on driver safety cannot be overstated. According to federal data, roughly 37,000 people died and another 2.7 million were injured in car accidents in the United States in 2017 alone.[1]


The United States Department of Transportation estimates that 94% of crashes involve driver-related factors, such as impaired driving, speeding or illegal maneuvers.[2] These striking numbers are a driving force behind the enthusiasm for autonomous vehicles, which are expected to substantially diminish or eliminate accidents caused by human error.


Such a shift would not only improve safety, but also transform the litigation landscape. Most obviously, a substantial reduction in accidents would likely produce a correspondingly substantial reduction in accident-related litigation. But commentators have suggested that the character of the remaining litigation would also likely shift.


If human factors are largely eliminated from car accidents, the supposition has been that liability for accidents would shift almost entirely to car manufacturers, removing or severely restricting the availability of comparative fault defenses.[3] Indeed, a number of automobile manufacturers have previously stated that they would assume liability for accidents involving self-driving technology in their automobiles, while also noting that the anticipated liability for such accidents is likely to be very small or nonexistent, due to the safety improvements that the technology is expected to confer.[4]


This approach also appears to reflect the assumption of potential regulators of autonomous vehicles. For example, in a white paper authored this month by the Governors Highway Safety Associationwhich is a nonprofit group that represents state highway safety officesa panel of experts commissioned by the association noted that while some have suggested that owners could remain liable for autonomous vehicles involved in accidents, [o]thers have suggested that the [autonomous vehicles] manufacturer [would] be responsible under product liability law.[5]


In reality, such statementswhile broadly correct, in noting that autonomous driving technology is likely to alter the shape of litigationlikely oversimplify things significantly. Already, what little litigation there has been over alleged defects in autonomous driving technology has involved allegations of fault by human parties to the accidents at issue, a trend that is likely to continue at least during the likely long transition period during which autonomous, semi-autonomous and nonautonomous vehicles coexist.[6]


Moreover, even assuming there is a shift away from driver liability, it is not clear that liability will shift entirely to automobile manufacturersespecially in light of the trend toward outsourcing the multiple necessary components for an autonomous driving system to a diverse array of third-party vendors. Computerization of the driving function also opens potentially new frontiers for liability that were previously confined to other fieldssuch as hacking, which traditionally has threatened informational rather than physical harm.


This article explores these three potential complications to the assumption that manufacturers necessarily would assume most or all liability for car accidents as autonomous vehicles are introduced.


Defining the scope of the duty to anticipate and prevent accidents


One question that complicates liability questions is what standard autonomous vehicles will have to meet in terms anticipating and preventing accidents. Needless to say, a relatively high standard of safety is likely to be imposed, especially in light of the expectation noted above that autonomous vehicles will ultimately prove to be much safer than traditional, human-driven automobiles.


But the law of product liability also generally recognizes that no product can be made completely safe, and (in many states) liability generally will not be imposed where a manufacturer has exercised reasonable care and the product meets state-of-the-art standards for safety. Thus, at least absent alteration of these ground rules by courts or legislatures seeking to set forth unique rules for self-driving cars (which is always a possibility), each accident holds the potential to raise the question whether the circumstances that led to the accident should have been anticipated and avoided under standards of reasonable care, or state of the art.


Both the advance of technology and the facts of the case will likely play heavily into this analysis. As to technology, the degree of vehicle autonomy will almost certainly be a relevant factor. Of note, vehicle autonomy is not strictly a yes/no question. The Society of Automotive Engineers has created a six-level scale, from zero to five, to classify how autonomous a vehicle is, with zero being entirely human-controlled in all situations, and five requiring no human input in any driving situation. This scale has been adopted by the National Highway Traffic Safety Administration.[7]


Level 0 is a traditional car with no automation, while Level 1 refers to limited automated safety features, such as crash-avoidance braking technology that is already available in many cars. Levels 2 and 3 transfer greater responsibility to the caracceleration and lane-centering at Level 2, for example, or full self-driving capability under certain circumstances at Level 3but still require monitoring and intervention by a human driver to ensure safety. Levels 4 and 5 represent true autonomyLevel 4 for normal driving conditions, and Level 5 for all driving conditions, including specialized applications such as off-road driving.[8]


Experience already indicates that assignment of fault for an accident will remain a difficult question at some degrees of semi-autonomous driving, especially Levels 2 and 3, which rely on continued driver attention, even though the driver is not necessarily expected to be making driving decisions on a continuous basis. In the case of the first person to die in a self-driving car, for example, the driver of a Tesla that collided with a tractor-trailer at 74 miles per hour was determined to be at fault by NHTSA for abdicating his responsibility to monitor the car while it was in self-driving mode.


As one article describing the accident explained, the Tesla system was only semi-autonomousit was intended to aid, not replace, human drivers.[9] But the driver had diverted his attention entirely from the driving task, and missed a seven-second window to avoid collision.[10] This fact suggests that the contributory or comparative fault of the driver in accidents involving semiautonomous vehicles will remain a live issue.


It appears that plaintiffs will attempt to circumnavigate these express limitations on the capabilities of semi-autonomous systems, by alleging that such systems lull drivers into a false sense of security. In Hudson v. Tesla Inc., for example, a Tesla driver alleged that the automated driving system on his car was defective because it failed to identify a car left parked on the highway, with which his car collided at full speed. Although the Tesla system at issue was the same semi-autonomous system described abovei.e., it was not fully autonomous and not intended to replace humans driversthe complaint alleged that the system or the marketing around it misleads drivers into believing they can rely upon the autopilot system … with minimal driver input.[11]


Similar theories have been raised in the past with respect to new automobile technologies. For example, a similar claim was raised in the context of keyless ignition systems, which differ from the transmission interlock systems of older cars that required a physical key to start the engine. At the time of that case, federal regulations mandated that manufacturers design cars in such a way that it was impossible to remove the key from the ignition unless the transmission was in the park setting, in order to prevent the vehicle from rolling away.


The car manufacturer had designed a car with an electronic key fob, obviating the need for a traditional ignition system in the wheel. Despite that, the car contained a holster in the wheel where the driver could place the electronic key fob while driving. But the car would still allow the driver to take the fob out of the holster when the engine was off, regardless of whether the car was in park.


The thrust of the plaintiffs argument was that this departure from the traditional physical transmission interlock system removed an important prompt to drivers to remember to put their car in park before exiting the vehicle. The theory was ultimately unsuccessful, because the court determined that the electronic key fob was not governed by federal regulations, and because any prudent person should know to put their car in park before exiting the car.[12] But it stands as an example of how litigation theories may shift as old-fashioned prompts for driver attention are phased out or replaced.


The facts of individual cases will also potentially shape the law on the scope of duty, even as to fully autonomous vehicles. Again, existing cases provide some clues about how these questions might arise. In Hudson v. Tesla Inc., noted above, the driver alleging that the automated driving system in his Tesla failed to identify a car left parked on the highway sued not only Tesla but also the owner of the parked car, claiming he was negligent in failing to properly remove his vehicle from the roadway.[13]


These allegations against two different defendantsthe car manufacturer and the owner of the parked vehicledirectly implicate the question of which party is at fault (or more at fault) in such circumstances. Put differently, it raises the issue whether (and to what extent) the manufacturer has a duty to account for the acts of third parties when those acts are unlawful. In this instance, the plaintiff alleged that an abandoned roadside vehicle is a foreseeable roadway hazard[], which offers one potential factor to consider in deciding such issues.[14]


But foreseeability becomes more difficult in other settings because it puts the focus on potentially very technical questions regarding what self-driving technology realistically is, or should be, capable of. In March 2018, for instance, a pedestrian was struck and killed by a self-driving car in Arizona, in what appears to be the first time an automobile driven by a computer has killed a human being by force of impact.[15] The ensuing investigation suggested that the pedestrian might have been at fault by stepping out from the shadows while jaywalking.[16]


Of course, jaywalking is also foreseeable in the broad sensebut even the fastest computer and the sharpest cameras may fail to identify a pedestrian who enters the roadway a split second before an approaching car speeds by. Similar scenarios involving collisions with other cars rather than pedestrians (either as a result of human error or other factors such as road conditions or wildlife entering the roadway) are easily imagined. This category of cases highlights that it will not be sensible to impose liability on manufacturers in every accident involving fully autonomous vehiclesit will depend on the facts.


In short, the few incidents of accidents involving self-driving cars already supply ample ground to conclude that a wholesale shift of liability from drivers to manufacturers is unlikely anytime soon, or perhaps ever. Liability will remain fact-bound, and depend on the type of technology involved, the warnings regarding its use and the actions of other partiesincluding drivers, as long as they still have a role to play, even if that role is reduced by the emergence of semi-autonomous driving features.


Identifying the relevant manufacturer


Another issue that might emerge in litigation against manufacturers of autonomous vehicle technology is the identification of the relevant manufacturer. Increasingly, automobile manufacturers have outsourced certain component parts to third-party manufacturers, and autonomous vehicle technology is no exception.


Fully autonomous vehicles require a number of systems working together in order to successfully see the road and drive the vehicle safely. Because autonomous vehicle technology comprises many constituent partswhich are likely to be provided by different suppliersthe issue of identifying the relevant manufacturer or manufacturers could be a complex matter, especially in light of the fact that the many of the components must work together in order to function.


Among other things, autonomous vehicle systems require processors (i.e., computer hardware), sensors of various sorts (e.g., cameras, radar and lidar, a sensor that operates like radar using lasers), communications systems, mapping systems (including failsafe systems that can continue to operate if communications are interrupted) and software algorithms (i.e., programs that synthesize all of the data from the sensors and the vehicle in order to safely drive it).[17]


These systems implicate different sets of expertise, including areas of expertise that have no prior application in the automobile spacewhich is driving suppliers such as semiconductor companies to enter the market, even if they did not previously have a presence in the automotive sector.[18] The result is that it is a near certainty that automobile manufacturers will not be designing or fabricating every aspect of these systems in house.[19]


The question thus arises: Which manufacturer would bear responsibility in the event of an accident resulting from alleged defects in an autonomous driving system? To some extent, such questions would not really be new. Auto accident litigation has already played host to cases involving alleged defects in components supplied by third parties, which are often named along with the automobile manufacturers as co-defendants.[20]


And here, too, the first lawsuits involving semi-autonomous vehicles suggest that this trend would continue. In Lommatzsch v. Tesla Inc., for example, the plaintiff sued after her Tesla collided with cars stopped on a highway, alleging that the autopilot system failed to brake. The plaintiff sued Tesla, but also sued a third-party service center that had replaced a sensor in the car prior to the crash, alleging that the co-defendant had negligently serviced the car.[21]


But the complexity and integrated nature of autonomous driving systems are likely to complicate these inquiries significantly. As noted above, these systems require numerous components that are likely to be supplied by numerous manufacturers. Such arrangements portend potentially difficult and complex causation issues in certain accident scenarios where locating (or excluding) the cause of failure in an integrated autonomous driving system is complicated and open to dispute.


Even governmental entities maintaining roadways may be subject to litigation under novel theories of liability. In one recent case in which a driver was killed when the car he was driving struck a crash attenuator on the side of a highway, the plaintiff sued not only the manufacturer but also the state of California, arguing that the crash attenuator was improperly maintained and contributed to the severity of the car accident.[22]


The allegations raise questions of what duty highway authorities may have to account for the emergence of self-driving vehicle technology in implementing traditional highway safety measures, and whether they have any duty to adopt new technologies to ensure the safe operation of such vehicles. In these respects, too, it may be premature to assume that liability will inevitably shift to automobile manufacturers in accidents alleged to result from failures in autonomous driving systems.


Dealing with hacking and informational failures


Finally, hacking and other informational failures also pose unique risks in the new world of autonomous vehicles. Autonomous driving systems vitally depend on information services in multiple formsfor example, in the mapping information and GPS data that inform navigation, and in the data such cars are expected to trade among each other in real time.


In particular, multiple companies have been experimenting with vehicle-to-vehicle, or V2V, communication, and the DOT has already proposed rules to make V2V communication mandatory in some situations.[23] Through V2V communication, vehicles could potentially send informationsuch as speed, position and directionto a cloud which would be available to all vehicles.[24] Other vehicles would then be able to read that data, using it to prevent crashes through coordination and advanced knowledge of incoming vehicles that a human driver would not necessarily be able to see.


While V2V communication has the potential to unclog traffic and reduce accidents, it also faces vulnerabilities from hackers, who could potentially read private passenger information, edit data to confuse autonomous vehicles or even send commands to vehicles on the road. Such developments implicate a potentially new frontier of liability for hacking and informational injurywhich in this context could have profound physical consequences.


One critical question would be the applicable legal doctrines.[25] Arguably, and depending on context, the data or software targeted by hacking might not be a product within the meaning of existing product liability law.[26] The answer to that question could have important consequences if, for example, a state had sought to limit or enhance product liability in certain situations, but hacking-caused accidents were determined to fall outside those legal reforms.


Whatever body of law would govern, hacking is also another area where identifying the appropriate defendant could prove difficult and fact-bound. Principal liability would presumably lie against the hacker, assuming one could be identified. The same component issues noted above could also be in play if there were a colorable claim that one or more components failed to provide needed data security.


A third-party provider of cloud services could also have exposure if the data breach occurred in its platformalthough its potential liability might not insulate component manufacturers from claims that autonomous driving systems should be able to detect and adjust for problems in V2V systems. Owners of the vehicles (who may increasingly be parties other than the passengers, as the profusion of autonomous vehicles is expected to be accompanied by a shift away from individual car ownership toward on-demand and ride-sharing services) could also face potential exposure in the event that they fail to update the vehicles software with security patches rolled out by the manufacturers.


In short, the unique reliance of autonomous vehicles on data suggests novel avenues of exposure in accidents, and these new roads may not all lead to the manufacturer of the car.


Conclusion


The expected safety advantages to be gained by a shift to autonomous vehicles are extensive, and will likely transform the litigation landscape relating to car accidents, which hopefully will be greatly diminished. As detailed above, however, the changing shape of litigation is unlikely to result in monolithic liability for automobile manufacturers in all cases. Rather, autonomous vehicle technology is poised to transform liability questions in multiple dimensions that are difficult to predict and will depend on factual circumstances.


Geoffrey M. Wyatt is a partner at Skadden Arps Slate Meagher & Flom LLP.


Paden Gallagher contributed significantly to the research and drafting of this article while employed as a summer associate at the firm.


The opinions expressed are those of the author(s) and do not necessarily reflect the views of the firm, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.

  1. Natl Highway Traffic Safety Admin., Traffic Safety Facts Annual Report, at 1 (Apr. 23, 2019), available at https://cdan.nhtsa.gov/tsftables/tsfar.htm.
  2. United States Dept. of Transp., Preparing for the Future of Transportation: Automated Vehicles 3.0, at 3 (Oct. 2018), available at https://www.transportation.gov/av/3.
  3. See, e.g., James M. Anderson et al., Autonomous Vehicle Technology: A Guide for Policymakers 11828 (RAND Corp. ed., 2016), available at https://www.rand.org/pubs/research_reports/RR443-2.html; David King, Putting the Reins on Autonomous Vehicle Liability: Why Horse Accidents Are the Best Common Law Analogy, 19 N. Carolina J. of L. & Tech. 127, 15658 (2018).
  4. E.g., Clifford Atiyeh, Volvo Will Take Responsibility If Its Self-Driving Cars Crash, Car & Driver, Oct. 8, 2015, https://www.caranddriver.com/news/a15352720/volvo-will-take-responsibility-if-itsself-driving-cars-crash/ (citing statements to this effect from Volvo, Mercedes-Benz and Google) .
  5. Governors Highway Safety Association, Automated Vehicle Safety Expert Panel: Engaging Drivers and Law Enforcement, Aug. 2019, https://www.ghsa.org/sites/default/files/2019-08/AV%20Safety%20White%20Paper_FINAL.pdf.
  6. So far at least four lawsuits have been filed against autonomous car manufacturers alleging products liability claims. Complaint for Damages, Nilsson v. Gen. Motors LLC, No. 3:18-cv-00471-KAW (N.D. Cal. Jan. 22, 2018); Complaint, Hudson v. Tesla Inc., 2018-CA-011812-O (Fla. Cir. Ct. Oct. 30, 2018); Complaint, Lommatzsch v. Tesla Inc., 2:18-cv-00775 (Utah D. Ct. Sept. 4, 2018); Complaint for Damages, Huang v. Tesla Inc., No. 19-cv-346663 (Super. Ct. Santa Clara County April 26, 2019). The Hudson matter expressly alleges fault against a third party for abandoning a car on the highway that was allegedly struck by the plaintiffs vehicle. See Hudson Compl. 44, 97. And in Nilsson, the plaintiff was a motorcyclist who alleged he was struck when an autonomous driving feature caused a vehicle manufactured by the defendant to suddenly switch lanes, but the defendant suggested in a statement to a state regulatory authority that the plaintiff put himself in harms way by driving between lanes of traffic. R.J. Vogt, GM Settles First-Known Suit Over Self-Driving Car-Crash, Law360, June 1, 2018, https://www.law360.com/articles/1049776/.
  7. United States Dept. of Transp., supra note 2, at vi; Hope Reese, Updated: Autonomous Driving Levels 0 To 5: Understanding The Differences, TechRepublic, Jan. 20, 2016, https://www.techrepublic.com/article/autonomous-driving-levels-0-to-5-understanding-thedifferences/.
  8. See, e.g., Reese, supra (summarizing and providing examples for the six levels) .
  9. Steven Seidenberg, Whos To Blame When Self-Driving Cars Crash?, ABA J., July 1, 2017, http://www.abajournal.com/magazine/article/selfdriving_liability_highly_automated_vehicle.
  10. Id.
  11. Nathan Hale, Tesla Sued Over Fla. Drivers Crash While Using Autopilot, Law360, Oct. 30, 2018, https://www.law360.com/articles/1097240/.
  12. Taragan v. Nissan N. Am. Inc. , No. C 09-3660 SBA, 2010 WL 3491000 (N.D. Cal. Sept. 2, 2010) (dismissing plaintiffs class claims which were based on the federal regulations) affd, 475 Fed. Appx 221 (9th Cir. 2012); Taragan v. Nissan N. Am. Inc. , No. C 09-3660 SBA, 2013 WL 3157918, at *9 (N.D. Cal. June 20, 2013) (dismissing plaintiffs claims, inter alia, the purported rollaway risk is completely avoidable by placing the automatic transmission in park and/or applying the parking brakeall of which a driver should do as a matter of common sense and prudence) .
  13. Hudson Compl. 44, 97.
  14. Hudson Compl. 3.
  15. Ian Bogost, Can You Sue A Robocar?, The Atlantic, Mar. 20, 2018, https://www.theatlantic.com/technology/archive/2018/03/can-you-sue-a-robocar/556007/.
  16. Id.
  17. Seunghyuk Choi et al., McKinsey & Co., Advanced Driver-Assistance Systems: Challenges And Opportunities Ahead, Feb. 2016, https://www.mckinsey.com/industries/semiconductors/ourinsights/advanced-driver-assistance-systems-challenges-and-opportunities-ahead.
  18. Id.
  19. See, e.g., Chuck Tannert, Could iPhone-Style Contract Manufacturing Come To The Car Industry?, Ars Technica, Aug. 13, 2018, https://arstechnica.com/cars/2018/08/could-iphone-stylecontract-manufacturing-come-to-the-car-industry/.
  20. Richard E. Kaye, American Law of Products Liability 8:9 (3d ed. 2019).
  21. Lommatzsch Compl. 12, 44-48.
  22. Huang Compl. 51-70.
  23. Hope Reese, US DOT Proposes V2V Mandate to Make Roads Safer, TechRepublic (Dec. 13, 2016, 8:51 AM), https://www.techrepublic.com/article/us-dots-proposed-rule-mandates-v2v-to-makeroads-safer-by-letting-cars-talk-to-each-other/; 6 Vehicle-to-Vehicle (V2V) Communication Startups, analyze (April 28, 2017), https://www.nanalyze.com/2017/04/6-vehicle-v2v-communicationstartups/.
  24. See Rick Blaisdell, Connected Cars and V2V Communication, Ricks Cloud (March 4, 2016), https://rickscloud.com/connected-cars-and-v2v-communication/.
  25. See Sunghyo Kim, Crashed Software: Assessing Product Liability For Software Defects In Automated Vehicles, 16 Duke L. & Tech. Rev. 300, 31112 (2018).
  26. Id.


Copyright 2019 Mondaq Ltd. All Rights Reserved.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.

Digital Workforce: A Game-Changer for All CFOs

The Malta Independent

August 29, 2019

How does reducing your cost by up to 75% while at the same time improving your work speed, accuracy, and control sound? At first, some may deem this to be too good to be true. However, the truth is robotic revolution is here, and advances in automation have actually made such an ideal scenario possible.


A golden opportunity for CFOs


Such advances have especially presented a valuable opportunity for Chief Financial Officers (CFOs) who in todays ever-changing environment are always seeking novel and innovative ways to reduce costs, achieve profitable growth, and drive competitive advantage. As bars get higher, they are challenged to think differently, and digital workforce is one of their answers.


Better efficiency through robotic process automation


The automation spectrum is wide-ranging, offering endless opportunities to those CFOs who strive to radically improve the efficiency of their finance department. On one end of the spectrum lies Robotic Process Automation (RPA), a kind of software that is able to carry out fixed, predictable, ruled-based tasks in areas like accounts reconciliation, transaction processing, and order entry.


It is true, robots have been around for a number of years now, having proved to be revolutionary in industries such as medicine, manufacturing, and warehouse operations. Now, robotics are morphing into digital workforce, and are said to be as disruptive to the back office as were outsourcing and offshoring.


In addition to reducing costs, robots also provide CFOs with a peace of mind, through their ability to reduce risk of manual error. Whats more, by transferring certain tasks to robots, employees can instead focus on work of higher value. Rather than spending an entire day repetitively processing transactions, employees can focus on examining customer profitability or pinpointing cost-saving opportunities. Thus, digital workforce becomes instrumental in increasing the employees capacity for more valuable business partnering.


Generally, a combination of people and technology is utilized to achieve business processes. Nowadays, given such a contemporary competitive environment, business needs are constantly changing to reach set targets and ensure maximum efficiency. Consequently, there might be technology limitations, as it is difficult for non-RPA based software applications to be adapted, evolved or developed at the same speed at which the business requires to effect these changes. This is an extremely good case where RPA can be a solution, since such automation has the capability and potential to cater for the ever-changing business needs.


Beyond RPAs


At the higher end of the automation spectrum, one finds RPA integrated with more refined, cognitive technologies like machine learning and natural language processing. These are capable of executing complex tasks that historically necessitated human intelligence and situational analysis. Such systems promise to assist employees in, amongst others, conducting research as well as in answering customers queries and crucial business questions.


As CFOs endeavor to reach new heights, robots, through their overwhelming combination of benefits, serve to vault them into position. And so, it comes as no surprise that the partnership between humans & machines and business & IT continues to evolve and thrive.


Savvy CFOs no longer aim to reduce costs through low-cost labor and geography. Rather, they are diverting their efforts to next-generation automation, which is undeniably causing unprecedented transformation in finance departments.


Copyright 2019 Standard Publications Ltd Provided by Syndigate.info, an Albawaba.com company All Rights Reserved.

Copyright © LexisNexis, a division of Reed Elsevier Inc. All rights reserved.  
Terms and Conditions   
Privacy Policy

Quality News Today is an ASQ member benefit offering quality related news
from around the world every business day.

Click here to read the entire article: Quality News Today.