Menu

Timesdelhi.com

June 25, 2019
Category archive

Autopilot

Consumer Reports knocks Tesla’s Navigate on Autopilot feature

in Automation/Automotive/Autopilot/consumer reports/Delhi/India/Politics/Tesla/tesla model 3/Tesla Model S/Transportation by

Consumer Reports is calling the automatic lane-change feature on Tesla’s Navigate on Autopilot “far less competent” than a human driver and cautioned it could pose safety risks.

The consumer advocacy organization posted its review Wednesday on the newest version of Tesla’s advanced driver assistance system.

Navigate on Autopilot is an active guidance system that is supposed to navigate a car from a highway on-ramp to off-ramp, including interchanges and making lane changes. Once drivers enter a destination into the navigation system, they can enable “Navigate on Autopilot” for that trip.

Tesla pushed out a software update last month to allow for automatic lane changes. Drivers have to enable this feature, which gives the car permission to make its own lane changes. If not enabled, the system asks the driver to confirm the lane change before moving over. Automatic lane changes can be canceled at any time.

The system has been touted as a way to make driving less stressful and improve safety. In practice, the system had startling behavior, Jake Fisher, senior director of auto testing at Consumer Reports told TechCrunch.

“It doesn’t take very long behind the wheel with this feature on to realize it’s not quite ready for prime time,” Fisher said. CR said one of the more troubling concerns were failures of Tesla’s three rearward-facing cameras to detect fast-approaching objects from the rear better than the average driver.

The CR reviewers found Navigate on Autopilot lagged behind human driving skills and engaged in problematic behavior such as cutting off cars and passing on the right. CR drivers often had to take over to prevent the system from making poor decisions.

As a result, the system increases stress and doesn’t improve safety, Fisher said, before asking “So what is the point of this feature?”

The automatic lane change reviewed by Consumer Reports is not the default setting for Autopilot, Tesla notes. It’s an option that requires drivers to remove the default setting. Tesla also argues that drivers using Navigate on Autopilot properly have successfully driven millions of miles and safely made millions of automated lane changes.

While Fisher acknowledged the default setting, he contends that isn’t the issue. He notes the Tesla has many warnings that the driver must be alert and ready to take over at any time.

“Our concern is that if you’re not alert (or ready to take over) you could be put into a tricky situation,” he said.

The bigger concern for all systems like these is the driver will put too much trust into it, Fisher said. The automatic lane-change feature might not be good enough for drivers to let down their guard yet. If Tesla improves this system, even a little bit, the risk of complacency and too much trust rises.

And that’s problematic because drivers still must be ready to take over. “Just watching automation is a harder human task than driving the car,” he said.

CR asserts that an effective driver monitoring system would mitigate this risk. DMS is typically a camera combined with software designed to track a driver’s attention and pick up on cognitive issues that could cause an accident such as drowsiness.

DMS are found in certain BMW models with an ADAS system called DriverAssist Plus, the new 2020 Subaru Outback and Cadillac’s equipped with its Super Cruise system.

This isn’t the first time CR has raised concerns about Autopilot. Last week, the consumer advocacy organization called on Tesla to restrict the use of Autopilot and install a more effective system to verify driver engagement in response to a preliminary report by National Transportation Safety Board on the fatal March 2019 crash of a Tesla Model 3 with a semi-trailer in Delray Beach, Fla.

Last year, CR gave GM’s Super Cruise the top spot in its first-ever ranking of partially automated driving systems because it is the best at striking a balance between technical capabilities and ensuring drivers are paying attention and operating the vehicle safely. Tesla followed in the ranking not because it was less capable, but because of its approach to safety, Fisher noted.

CR evaluated four systems: Super Cruise on the Cadillac CT6, Autopilot on Tesla Model S, X and 3 models, ProPilot Assist on Infiniti QX50 and Nissan Leaf, and Pilot Assist on Volvo XC40 and XC60 vehicles. The organization said it picked these systems because they’re considered the most capable and well-known in the industry.

Nvidia dives into a new business segment with Drive AutoPilot

in Artificial Intelligence/Automation/Automotive/Autopilot/CES 2019/continental/deep neural networks/Delhi/drive/driver/Emerging-Technologies/India/nvidia/Politics/self-driving car/Technology/Tesla/transport/Transportation/Xavier/zf by

Nvidia’s automotive ambitions seemed targeted solely on creating a platform to enable fully autonomous vehicles, notably the robotaxis that so many companies hope to deploy in the coming decade.

It turns out that Nvidia has also been working a more near-term product that opens it up to a different segment in the automotive industry. The company announced Monday at CES 2019 that it has launched Nvidia Drive AutoPilot, a reference platform that automakers can use to bring more sophisticated automated driving features into their production vehicles. This is not a self-driving car product, although it will likely be misinterpreted as such.

The Drive AutoPilot system is meant to make those advanced driver assistance system in today’s cars even better. It enables highway merging, lane changes, lane splits, pedestrian and cyclist detection, parking assist, and personal mapping as well as in-cabin features like driver monitoring, AI copilot capabilities, and advanced in-cabin visualization of the vehicle’s computer vision system. It also allows for over-the-air software updates, a capability that automakers, with the exception of Tesla, have been slow to adopt.

Nvidia already has two customers for Nvidia Drive AutoPilot — a name not to be confused with Tesla’s consumer-facing semi-autonomous system Autopilot. (It should be noted that a Tesla uses a derivative of Nvidia’s Drive platform, although that could change. Tesla has been developing its own chip, otherwise known as “Hardware 3.”

On Monday, Tier 1 suppliers Continental and ZF announced that by 2020 they will have partially automated driving systems ready for production that are based on the Drive AutoPilot platform.

Nvidia argues that there’s a market for this improved automation, noting a recent Insurance Institute for Highway Safety study that found existing Level 2 ADAS systems “offer inconsistent vehicle detections and poor ability to stay within lanes on curvy or hilly roads, resulting in a high occurrence of system disengagements where the driver abruptly had to take control.” Level 2 is a designation from SAE that means the vehicle’s automated system can handle accelerating, braking and steering, but must still be monitored by the driver, who should be prepared to take control at any time.

Nvidia doesn’t make plug-and-play type systems. Instead, Continental, ZF or other suppliers can take this reference platformand use it to deliver any combination of more advanced automation. For example, Continental will use it to produce an automated driving and parking solution that will be available to customers by 2020.

The foundation of the Drive AutoPilot is Nvidia’s Xavier system-on-a-chip processor, which can handle some 30 trillion operations per second. Then it adds Nvidia’s Drive software to process deep neural networks for perception as well as data pouring in from surround camera sensors.

The Drive AutoPilot system is part of Nvidia’s broader Drive platform. It’s also designed to complement the company’s Nvidia Drive AGX Pegasus system that provides Level 5 capabilities for robotaxis.

 

“A full-featured, Level 2+ system requires significantly more computational horsepower and sophisticated software than what is on the road today,” Rob Csongor, vice president of Autonomous Machines at Nvidia said, adding that the company’s system makes it possible for carmakers to quickly deploy advanced autonomous solutions by 2020 and to scale this solution to higher levels of autonomy faster.

Tesla Model X sped up in Autopilot mode seconds before fatal crash, according to NTSB

in Automotive/Autopilot/Delhi/India/Politics/TC/Tesla/Tesla Model x/Transportation by

The National Transportation Safety Board has released a preliminary report detailing the fatal crash involving a Tesla Model X in March. The crash also resulted in a fire and shut down two lanes of Highway 101 near Mountain View, Calif. At this point, the NTSB has yet to determine a probable cause of the crash and is continuing to investigate the accident.

The report says the Model X, while in Autopilot mode, sped up to 71 mph in the seconds leading up to the crash.

“At 3 seconds prior to the crash and up to the time of impact with the crash attenuator, the Tesla’s speed increased from 62 to 70.8 mph, with no precrash braking or evasive steering movement detected,” the report states.

Source: NTSB/S. Engleman

Tesla’s Autopilot mode is designed to match the speed of a slower vehicle traveling ahead of it. At the time, Autopilot was set to 75 mph, according to the NTSB.

A Tesla spokesperson declined to comment but pointed me to its March blog post, where the company describes how the driver’s hands were not detected on the wheel for the six seconds prior to the collision. The NTSB confirmed that in its preliminary report today.

“Tesla Autopilot does not prevent all accidents – such a standard would be impossible – but it makes them much less likely to occur,” Tesla wrote in a March blog post. “It unequivocally makes the world safer for the vehicle occupants, pedestrians and cyclists.”

Walter Huang, the owner of the car who died as a result of the crash, had previously taken his car into the Tesla dealership, saying his car had a way of veering toward the exact barrier his car hit, ABC7 reported. Tesla, however, previously said it has no record of Huang complaining about Autopilot.

Tesla makes Autopilot easier to use in the Model 3

in automotive industry/Autopilot/Delhi/India/Politics/steering wheel/TC/Tesla/Tesla Autopilot/transport by

The Model 3 now has a new way to control Autopilot. The company recently sent out an update that moved the controls from the center infotainment stack to the steering wheeling. Previously, drivers had to use the large screen to change Autopilot’s speed and cruise distance and in doing so, required drivers to take their eyes off the road to change these options.

The Model 3 is an exercise in minimalism, and to that end, the company is seemingly still working out the best interface. To that end, the controls on the Model 3’s steering wheel were purposely not labeled or dedicated to a specific function and instead change depending on the car’s role.

Thanks to the 2018.12 update, the right-hand scroll wheel controls the speed of the vehicle while the buttons flanking that wheel changes the follow distance. The new controls do not replace the existing controls but rather supplement them.

During our review of the Model 3, we noted the sparse cockpit design and found it good and bad. On the one hand, it’s terrific to have an unobstructed view of the road – it’s as pure as driving experience as rolling down the track in the soapbox derby car of your youth, and it leaves you feeling connected to the road itself. But yet the car feels too reliant on the center touchscreen, which often requires drivers to look away from the road for simple commands.

Tesla is operating under increasing scrutiny following a fatal accident that involved Autopilot. Updates like this show the company is at least listening to owners and updating its vehicles in kind.

Tesla said to be working on its own self-driving AI chip with AMD

in amd/Artificial Intelligence/Automotive/autonomous driving/Autopilot/Delhi/India/Politics/self-driving/TC/Tesla/Transportation by

Tesla is reportedly developing its own processor for artificial intelligence, intended for use with its self-driving systems, in partnership with AMD. Tesla has an existing relationship with Nvidia, whose GPUs power its Autopilot system, but this new in-house chip reported by CNBC could potentially reduce its reliance on third-party AI processing hardware.

AMD would help provide Tesla with development of the chip. The electric carmaker has apparently already received the first samples of its processor prototype for testing, according to the report. Building a specific chip designed entirely for autonomous driving would help Tesla in a number of ways, including gaining more control over a key aspect of its own supply chain, and increasing overall system efficiency and performance.

Tesla has claimed that its current generation vehicles have all the hardware and computing capability on board to achieve full self-driving with a software update in the future. But the company is also always iterating on its hardware, eschewing the traditional model year upgrades of legacy automakers.

Tesla’s autopilot business is run by Jim Keller, who previously spent much of his career at AMD, and then at Apple, where he helped design the Apple A4 and A5 in-house processors. Keller’s key role at the company would likely have influenced the automaker to drive towards owning its own processor design to more closely align its self-driving software and compute hardware components.

Go to Top