Connect with us

Published

on

NHTSA is asking Tesla to do what it has always been able to avoid: release data from its ‘Full Self-Driving (FSD)’ program.

The agency wants to know how closely its planned robotaxi service in Austin will be to its FSD program, which is currently under investigation for safety defects.

NHTSA, the agency in charge of automobile safety regulations in the US, appears to know very little about Tesla’s planned rollout of a “robotaxi service” in Austin, Texas, even though it is reportedly just a few weeks away.

The agency is currently investigating Tesla’s ‘Supervised Full Self-Driving’, FSD, program and when it heard from Tesla recently that the planned robotaxi service in Austin is going to be based on its FSD program, it got worried.

Advertisement – scroll for more content

Tanya Topka, Director of NHTSA’s Office of Defects Investigation, wrote to Tesla in an email sent last week and obtained by Electrek:

As you are aware, NHTSA has an ongoing defect investigation (PE24031) into FSD collisions in reduced roadway visibility conditions.

They are asking Tesla for more information:

The agency would like to gather additional information about Tesla’s development of technologies for use in “robotaxi” vehicles to understand how Tesla plans to evaluate its vehicles and driving automation technologies for use on public roads.

I included the entire list of questions below, but in short, NHTSA wants to know what parts of Tesla’s ‘Supervised FSD’, which is under safety investigation, will be used in the robotaxi service planned for Austin and other expansions later this year, and if it is different, what the differences are.

They want to know what data Tesla has to prove its vehicles are safe enough to be fully autonomous in this paid robotaxi service.

As we have often reported before, Tesla famously has avoided ever releasing any data about its FSD program. We have to rely on crowdsourced data to track progress in the program.

The automaker also avoids releasing critical data to NHTSA in its ADAS crash reporting system.

Tesla has until June 19 to respond or face up to $27,874 in penalties per violation per day.

Here’s the complete list of questions NHTSA is asking Tesla about its planned rollout of a robotaxi service in the US:

  1. Based on Tesla’s public statements described above, NHTSA understands that Tesla is developing an automated driving system (ADS) based on its current FSD Supervised system, which Tesla has labeled an advanced driver assistance system. State the name(s) of the system(s) that will be used in robotaxi development and deployment as well as Tesla’s position on the SAE Level classification for the purposes of reporting under NHTSA’s Standing General Order on crash reporting.
  2. Describe Tesla’s plans to develop, test, and commercialize a robotaxi or analogous technologies on public roadways, including details regarding:
    • The number of vehicles by make and model anticipated at start of on-road operations and within the subsequent 12 and 24 months.
    • To the extent that Tesla plans to use any new vehicle models in the next 24 months, explain whether any vehicles that do not fully comply with Federal Motor Vehicle Safety Standards (49 C.F.R. Part 571) will be operated on public roadways, and state whether Tesla plans to seek any FMVSS exemptions.
    • The expected timetable for availability of a robotaxi or similar service to the public or other groups.
    • The expected timetable for availability of robotaxi technology for operation on vehicles controlled by people or entities other than Tesla and whether Tesla will require such individuals or entities to meet certain requirements to ensure safe operations.
    • The locations anticipated at start of on-road operations and within the subsequent 12 and 24 months.
    • Whether and how vehicles will be supervised or otherwise monitored by Tesla in real time.
    • Use of any teleoperation technologies such as remote driving and remote assistance and the limits of control authority for remote input to system/vehicle operation.
    • The roles and responsibilities of any in-vehicle or remote staff involved in monitoring, supervising, or intervening in system operation.
  3. Describe the driving automation system(s) that will be used for the robotaxi effort and any relationship to Tesla’s existing FSD Supervised product available to consumers today.
    • Descriptions of each perception sensor (including count and location), each compute subsystem, and overall system architecture for perception, planning, actuation, and performance monitoring/logging.
    • The role(s) of any cameras or other sensors within the vehicle cabin for the robotaxi system’s safe operation when supervised and unsupervised.
    • Explanations of differences in system implementation for the robotaxi and FSD Supervised.
    • Describe the maximum control authority for the system when engaged including commanded speed, acceleration, braking, steering angle, permissible gear selection states while engaged, and limits on specialized maneuvers (e.g., reversing, parking, etc.).
    • Description of whether Tesla complies fully or partially with any industry standards, best practices, or guidance for the development and safety assurance of driving automation systems (e.g., SAE J3018, ISO/TS 16949, ISO 26262, SOTIF, UL4600, etc.).
  4. Provide a detailed description of the operational design domain (ODD) for the robotaxi driving automation system, including an explanation of:
    • ODD elements3 and associated thresholds for the ODD for each automation feature.
    • The set of ODD elements that are monitored by the automation system.
    • The set of ODD elements that are solely monitored by any in-vehicle or remote staff.
    • The designed response of the automation feature, for each ODD element, if a system limit is exceeded or an ODD exit occurs.
    • Specific operational restrictions Tesla is implementing (e.g., relating to time-of-day, weather, geofencing, maximum speed) and whether each operational restriction is implemented primarily to ensure safe operations within the subject system’s ODD.
  5. Describe how Tesla plans to determine whether its robotaxi system has achieved acceptably safe behavioral competency for a given ODD scope including:
    • a. Establishing behavioral competency thresholds for supervised on-road operations.
    • Establishing behavioral competency thresholds for on-road operations without real-time supervision.
    • Determining which behavioral competencies (and associated ODD elements) do not satisfy established thresholds for on-road operations both with and without real-time supervision.
    • How this approach aligns with or differs from Tesla’s processes for FSD Supervised.
  6. Explain Tesla’s approach for monitoring in-use interventions for the robotaxi system as it relates to:
    • Defining and tracking the types and frequency of disengagements or other human interventions – including both in-vehicle and remote interventions – and their relationship to safe driving behaviors.
    • Provide the current metrics for disengagements/interventions for the robotaxi system.
    • Planned differences in monitoring disengagements/interventions in comparison to Autopilot and FSD Supervised.
  7. Describe Tesla’s design and approach for emergency scenarios including:
    • Crash detection and response, including adequacy of minimal risk conditions depending on crash scenario.
    • The designed/intended maneuvers and/or other responses to achieve a stable stopped condition – i.e., a minimal/mitigated risk condition (MRC) – or takeover following a crash, system failure, ODD exit, or other scenario requiring an appropriate disengagement or other intervention.
    • Planned operational steps following achievement of an MRC.
    • Subject system and subject vehicle interactions with first responders.
    • Tesla’s operational response to incidents occurring with the subject system.
  8. Explain the methods and processes (e.g., establishing a safety case) in detail that are employed by Tesla to determine readiness of the robotaxi system for on-road use with and without supervision.
    • Explain whether Tesla employs a safety case or similar methodology. Describe how Tesla gathers and assesses evidence that its robotaxi system is ready for onroad use under supervision and without supervision.
    • List all processes Tesla has established for internal decision making on whether the system is acceptably safe for on-road use (e.g., satisfying whether safety claims in a safety case have been fully satisfied). Identify the accountable decision makers by name, role, and organizational structure.
    • List and describe each process that Tesla uses to establish metrics and associated baselines or thresholds that quantify acceptable performance for on-road use. Include descriptions of how the metrics are established.
    • Describe Tesla’s verification and validation methodology for the robotaxi product for metrics/thresholds including:
      • How Tesla identifies and handles potential performance gaps and regressions during development and while in use.
      • How changes or updates to existing metrics or thresholds are approved.
      • Tesla’s use of simulation, test track, and on-road testing as well as whether Tesla is leveraging data from consumer owned vehicles for verification or validation efforts of the robotaxi product.
    • To the extent that Tesla is using performance thresholds or metrics established based on human drivers, identify the source of the underlying data Tesla is using to establish the thresholds/metrics.
    • Explain how the system is designed to comply with traffic safety laws and how Tesla will monitor for compliance with traffic safety laws including traffic control devices, interactions with construction zones, and interactions with first responders.
    • Describe Tesla’s plan to collect, evaluate, and retain data to continuously monitor the ongoing operational performance metrics/thresholds.
    • Describe Tesla’s approach for determining if an operational performance metric/threshold has been violated.
  9. Describe how Tesla intends to ensure the safety of its robotaxi operations in reduced roadway visibility conditions, such as sun glare, fog, airborne dust, rain, or snow. In your response, describe whether Tesla’s approach differs, if at all, for a ride in which the reduced roadway visibility condition exists at the beginning of the ride and a ride in which the reduced roadway visibility condition first appears or is encountered during a ride.

FTC: We use income earning auto affiliate links. More.

Continue Reading

Environment

China installs the world’s most powerful wind turbine

Published

on

By

China installs the world's most powerful wind turbine

China’s Dongfang Electric has installed a 26-megawatt offshore wind turbine, snatching the title of world’s most powerful from Siemens Gamesa’s 21.5 turbine in Denmark.

Photo: Dongfang Electric Corporation

The Chinese state-owned manufacturer announced today that it has installed the world’s most powerful wind turbine prototype at a testing and certification base. This turbine, the world’s largest for capacity and size, boasts a blade wheel diameter of more than 310 meters (1,107 feet) and a hub height of 185 meters (607 feet). Dongfang shipped the turbine’s nacelle earlier this month – the world’s heaviest – along with three blades.

This offshore wind turbine is designed for areas with wind speeds of 8 meters per second and above. With average winds of 10 meters per second, just one of these giants can generate 100 GWh of power annually, which is enough to power 55,000 homes. That’s enough to cut standard coal consumption by 30,000 tons and reduce CO2 emissions by 80,000 tons. Dongfang says it’s wind resistant up to 17 (200 km/h) on the extended Beaufort scale.

In May, Dongfang said it had completed static load testing on the turbine’s blades, and the turbine is now undergoing fatigue testing, which could take up to a year before the turbine is fully certified.

Advertisement – scroll for more content

Read more: Trump just killed all offshore wind zones as US power needs surge


The 30% federal solar tax credit is ending this year. If you’ve ever considered going solar, now’s the time to act. To make sure you find a trusted, reliable solar installer near you that offers competitive pricing, check out EnergySage, a free service that makes it easy for you to go solar. It has hundreds of pre-vetted solar installers competing for your business, ensuring you get high-quality solutions and save 20-30% compared to going it alone. Plus, it’s free to use, and you won’t get sales calls until you select an installer and share your phone number with them. 

Your personalized solar quotes are easy to compare online and you’ll get access to unbiased Energy Advisors to help you every step of the way. Get started here.

FTC: We use income earning auto affiliate links. More.

Continue Reading

Environment

John Deere joins the robot revolution with GUSS acquisition

Published

on

By

John Deere joins the robot revolution with GUSS acquisition

The autonomous ag equipment experts behind the GUSS robotic sprayers have been developing their AI tech as part of a JV with John Deere for years — and now, that marriage is official. John Deere has acquired 100% of GUSS, and has big plans to pick up that tech and run with it like a … well, you know.

The latest battery-powered GUSS autonomous sprayer made its debut at the 2024 World Ag Expo show in Tulare, California, last summer, where executives from Deere called it, “the world’s first and only fully electric autonomous herbicide orchard sprayer.”

Since then, interest in automated ag equipment has only grown — fueled not just by rising demand for affordable food and produce, but by a national labor shortage made worse by the Trump Administration’s tough anti-immigration policies as well. It’s specifically those challenges around labor availability, input costs, and crop protection that GUSS and John Deere have been spending millions to address.

“Fully integrating GUSS into the John Deere portfolio is a continuation of our dedication to serving high-value crop customers with advanced, scalable technologies to help them do more with less,” explains Julien Le Vely, director, Production Systems, High Value & Small Acre Crops, at John Deere. “GUSS brings a proven solution to a fast-growing segment of agriculture, and its team has a deep understanding of customer needs in orchards and vineyards. We’re excited to have them fully part of the John Deere team.”

Advertisement – scroll for more content

About GUSS


GUSS autonomous farm sprayer; via John Deere.
GUSS autonomous farm sprayer; via John Deere.

The GUSS electric sprayer is powered by a Kreisel Battery Pack 63 (KBP63), which has a nominal energy capacity of 63 kWh, enabling the machine to operate for 10-12 continuous hours between overnight (L2) charges.

The GUSS electric sprayers feature the Smart Apply weed detection system that measures chlorophyll in the various plants it encounters, identifying weeds embedded among the crops, and only sprays where weeds are detected. The company claims its weed detecting tech significantly reduces the amount of chemicals being sprayed onto farmers’ crops, resulting in “up to 90% savings” in sprayed material.

John Deere’s deep pockets will support GUSS as it continues to expand its global reach, and help the group to accelerate Smart Apply’s innovation and integration with other John Deere precision agriculture technologies.

“Joining John Deere enables us to tap into their unmatched innovative capabilities in precision agriculture technologies to bring our solutions to more growers around the world,” says Gary Thompson, GUSS’ COO. “Our team is passionate about helping high-value crop growers increase their efficiency and productivity in their operations, and together with John Deere, we will have the ability to have an even greater impact.”

GUSS-brand autonomous sprayers will be sold and serviced exclusively through John Deere dealers, and the GUSS business will retain its name, branding, employees, and independent manufacturing facility in Kingsburg, California.

More than 250 GUSS machines have been deployed globally, having sprayed more than 2.6 million acres over 500,000 autonomous hours of operation.

Electrek’s Take


John Deere and GUSS Automation Unveil Electric Option and Smart Apply Upgrade

Population growth, while slowing, is still very much a thing – and fewer and fewer people seem to be willing to do the work of growing the food that more and more people need to eat and live. This autonomous tech multiplies the efforts of the farmers that do show up for work every day, and the fact that it’s more sustainable from both a fuel perspective and a toxic chemical perspective makes GUSS a winner.

SOURCE I IMAGES: John Deere.


If you’re considering going solar, it’s always a good idea to get quotes from a few installers. To make sure you find a trusted, reliable solar installer near you that offers competitive pricing, check out EnergySage, a free service that makes it easy for you to go solar. It has hundreds of pre-vetted solar installers competing for your business, ensuring you get high-quality solutions and save 20-30% compared to going it alone. Plus, it’s free to use, and you won’t get sales calls until you select an installer and share your phone number with them. 

Your personalized solar quotes are easy to compare online and you’ll get access to unbiased Energy Advisors to help you every step of the way. Get started here.

FTC: We use income earning auto affiliate links. More.

Continue Reading

Environment

Tesla asks court to throw out $243 million verdict in fatal Autopilot crash case

Published

on

By

Tesla asks court to throw out 3 million verdict in fatal Autopilot crash case

Lawyers for Tesla filed a motion asking a court to throw out a recent $243 million verdict against the company related to a fatal crash in Florida in 2019. The case is the first instance of Tesla being ruled against by a court in an Autopilot liability case – previous cases had ended up settled out of court.

To catch up, the case in question is the $243 million Autopilot wrongful death case which concluded early this month. It was the first actual trial verdict against the company in an Autopilot wrongful death case – not counting previous out-of-court settlements.

The case centered around a 2019 crash of a Model S in Florida, where the driver dropped his phone and while he was picking it up, the Model S drove through a stop sign at a T-intersection, crashing into a parked Chevy Tahoe which then struck two pedestrians, killing one and seriously injuring the other.

Tesla was also caught withholding data in the case, which is not a good look.

Advertisement – scroll for more content

In the end, for the purposes of compensatory damages, the driver was found 67% responsible and Tesla was found 33% responsible. But Tesla was also slapped with $200 million in punitive damages. The plaintiffs reached a settlement with the driver separately.

Tesla said at the time that it planned to appeal the case, and its first move in that respect happened today, with lawyers for Tesla filing a 71-page motion laying out the problems they had with the trial.

In it, Tesla requests either that the previous verdict be thrown out, that the amount of damages be reduced or eliminated, or that the case go to a new trial, based on what Tesla contends were numerous errors of law during the trial.

The table of contents of Tesla’s filing lays out the company’s rough arguments for why it’s requesting the verdict to be thrown out, with Tesla seeming to throw several arguments at the wall to see what sticks:

  • I. Tesla Is Entitled to Judgment as a Matter of Law (or at Least a New Trial) on Liability.
    • A. The Verdict Is Unsupported by Reliable Expert Evidence.
    • B. Plaintiffs’ Design-Defect Theories Fail as a Matter of Law.
      • 1. Tesla’s 2019 Model S Was Not Defective.
      • 2. McGee Was the Sole Cause of Plaintiffs’ Injuries.
    • C. The Failure-to-Warn Claim Fails as a Matter of Law.
      • 1. Tesla Had No Duty to Warn.
      • 2. Tesla Provided Extensive Warnings.
      • 3. The Asserted Failure to Warn Didn’t Cause the Crash.
    • D. Tesla Is Entitled to a New Trial If the Record Cannot Sustain the Verdict as to Any Theory on Which the Jury Was Instructed.
  • II. Highly Prejudicial Evidentiary Errors Warrant a New Trial on All Issues.
    • A. The Improper Admission of Data-Related Evidence Prejudiced Tesla.
    • B. The Improper Admission of Elon Musk’s Statements Prejudiced Tesla.
    • C. The Improper Admission of Dissimilar Accidents Prejudiced Tesla.
  • III. This Court Should Grant Tesla Judgment as a Matter of Law on Punitive Damages or at Least Significantly Reduce Punitive Damages.
    • A. Florida Law Prohibits the Imposition of Any Punitive Damages in This Case.
    • B. Florida Law Caps Punitive Damages at Three Times the Compensatory Damages Actually Awarded Against Tesla.
    • C. The Due Process Clause Limits Punitive Damages Here to No More Than the Net Award of Compensatory Damages.
      • 1. Tesla’s Conduct Was Not Reprehensible.
      • 2. A Substantial Disparity Exists Between the $200 Million Award of Punitive Damages and the $42.3 Million Award of Compensatory Damages.
      • 3. Comparable Civil Penalties Do Not Justify the Punitive-Damages Award.
  • IV. This Court Should Reduce the Grossly Excessive Award of Compensatory Damages to No More Than $69 Million.

In short, Tesla blames the driver (who was found 67% liable) fully for the crash, says that the Model S and its Autopilot system were state-of-the-art and not defective because “no car in the world at the time” could have avoided the accident, that it provided proper warnings even though it didn’t need to, that evidence was improperly admitted to prejudice the jury against Tesla, and that the punitive damages are excessive.

After looking through the document, Tesla’s main contention seems to be with the admission of various evidence that it says prejudiced the jury against Tesla.

Indeed, the only exhibit attached to the filing is a transcript of a podcast episode where one of plaintiffs’ experts talks about evidence that Tesla withheld data, which Tesla says should have been inadmissible and prejudiced the jury against it.

The plaintiffs repeatedly asserted that Tesla had deliberately withheld or tried to delete data, which required them to bring in third party experts to discover and examine the data.

Tesla says that the only reason these arguments were brought into court was to make the jury feel like there was a coverup, even though Tesla claims that there was no coverup. By repeatedly mentioning this, Tesla says the jury had a more negative view of the company than was fair.

It also says that Tesla CEO Elon Musk’s statements about Autopilot shouldn’t have been admissible, and that they prejudiced the jury against Tesla. Tesla says that the statements by Musk shown at the trial were irrelevant to plaintiffs’ case, exceeded the limits the court had set on which statements would be admissible, and that the admission of these statements “would disincentivize companies from making visionary projections about anticipated technological breakthroughs.”

You can read through the full filing here.

Update: After this story was published, plaintiffs’ attorneys reached out with their own statement

“This motion is the latest example of Tesla and Musk’s complete disregard for the human cost of their defective technology. The jury heard all the facts and came to the right conclusion that this was a case of shared responsibility, but that does not discount the integral role Autopilot and the company’s misrepresentations of its capabilities played in the crash that killed Naibel and permanently injured Dillon. We are confident the court will uphold this verdict, which serves not as an indictment of the autonomous vehicle industry, but of Tesla’s reckless and unsafe development and deployment of its Autopilot system.”  

Brett Schreiber of Singleton Schreiber, lead trial counsel for plaintiffs Dillon Angulo & Naibel Benavides.

Electrek’s Take

Reading through the filing is persuasive at first, but remember that this is only one side of the story – and Tesla is well-known for never budging an inch in legal or reputational matters. (Update: for a quick reaction from “the other side,” see the statement by plaintiffs’ attorneys directly above).

Thinking a little deeper, the filing does rely on a similar “puffery” argument which Tesla has used before. The idea here is that Musk’s statements should be ignored because he, as the CEO of the company, has an incentive (and well-known tendency) to overstate the capabilities of its vehicles.

Lawyers did not use that exact word here, but they do claim that Musk’s statements are “forward-looking” and “visionary.”

But, for a guy who talks so much that he wasted $44 billion on a $12 billion social media site (twice) so that he could force his words in front of every user every day, denying that his words have an effect is a strange legal argument.

Indeed, Tesla has a history of not doing paid advertisements in traditional media, and has relied on Musk, and specifically Musk’s twitter account, to be the company’s impromptu communications platform. Musk even closed the company’s PR department, instead taking on the full burden of that himself.

So to argue that Musk’s statements shouldn’t be admissible, or that they didn’t set the tone for the organization, is more than a little silly.

While Tesla and Musk did state many times that Autopilot was not full self-driving (although, neither was the feature they marketed under the name, ahem, “Full Self-Driving”), the balance of Musk’s statements describing Tesla’s features definitely could have led a driver to think that the vehicles were more capable than any other vehicle on the road.

This is why it’s strange that Tesla also argues that “no other car” could have stopped in the situation of the crash. If your company is constantly claiming that you have the best, safest, most autonomy-enabled vehicle in the world (including in this filing, where it is referred to as “state of the art”), then who cares whether other cars could have done it or not? We’re talking about your car, not anything else.

Further, Tesla said that admitting these statements will put a chilling effect on every corporation’s ability to project anticipated breakthroughs in tech. To this I say, frankly: good. Enough with the nonsense, lets focus on reality, and lets stop excusing lies as corporate puffery, across all industries.

But this is an example of Tesla trying to have it both ways, to pretend that Musk’s statements are just puffery but also that they are important to breakthroughs and that silencing Musk would harm the company. Yes, it probably would harm Tesla’s outreach – because Musk’s statements are roughly the only source of Tesla’s advertising, which is why they ought to be heard to establish what the public thinks about the capabilities of Teslas.

And while Tesla says that cases like these would “chill” development of safety features if manufacturers are punished for bringing them to market, the punishment here isn’t for bringing the feature to market, it’s for overselling the feature in a way that set public expectations too high. Other features have not received this sort of scrutiny because other features don’t get pumped up daily with ridiculous overstatements by the company’s sole source of advertising.

On the other points, I’m not a lawyer. I’m not up to date on the specific limits to punitive damages in Florida. But on the surface, it seems fair to me that if a company was found to withhold data in an important case, after declining a settlement, that some level of significant punishment is fair.

After all, withholding data in a single non-fatal crash that wasn’t even their fault is what led Cruise to shut down operations everywhere. That may have been an overreaction and would certainly be an overreaction in this case with Tesla, given the driver’s responsibility for the crash. But in this case, the damage done to people (a death) was greater, and the damages Tesla is being told to pay ($243 million) will not lead to a shutdown of the entire company. Especially considering this is the same company that just managed to find tens of billions of dollars to give to a bad CEO.


The 30% federal solar tax credit is ending this year. If you’ve ever considered going solar, now’s the time to act. To make sure you find a trusted, reliable solar installer near you that offers competitive pricing, check out EnergySage, a free service that makes it easy for you to go solar. It has hundreds of pre-vetted solar installers competing for your business, ensuring you get high-quality solutions and save 20-30% compared to going it alone. Plus, it’s free to use, and you won’t get sales calls until you select an installer and share your phone number with them.

Your personalized solar quotes are easy to compare online and you’ll get access to unbiased Energy Advisors to help you every step of the way. Get started here.

FTC: We use income earning auto affiliate links. More.

Continue Reading

Trending