Off the Kuff Rotating Header Image

NTSB

Tesla crash update

Interesting.

Federal investigators on Monday released their preliminary assessment of a deadly crash in Spring involving a Tesla, though many details of the incident remain under scrutiny, including whether anyone was in the driver’s seat at the time of the wreck.

In the two-page update, the National Transportation Safety Board said the crash occurred less than 600 feet from where Dr. William Varner, 59, and Everette Talbot, 69, began their trip in Varner’s driveway. NTSB officials said security camera footage shows Varner getting into the driver’s seat and Talbot getting in the front passenger seat.

Both were killed in the fiery crash, which took firefighters hours to extinguish because the car’s battery reignited several times.

Where Varner was when the crash happened is a point of contention between local crash investigators and Tesla officials. In the days after the crash, Harris County Precinct 4 Constable Mark Herman said investigators are confident no one was in the driver’s seat.

Tesla founder Elon Musk and other company officials have disputed that, saying evidence points to someone being in the seat, based on the damage to the steering column.

NTSB investigators in the report do not make a final determination.

“The car’s restraint control module, which can record data associated with vehicle speed, belt status, acceleration, and airbag deployment, was recovered but sustained fire damage,” investigators said in the report. “The restraint control module was taken to the National Transportation Safety Board recorder laboratory for evaluation.”

See here and here for the background, and here for a copy of the NTSB report. The key bit is this:

The vehicle was equipped with Autopilot, Tesla’s advanced driver assistance system. Using Autopilot requires both the Traffic Aware Cruise Control and the Autosteer systems to be engaged.[2] NTSB tests of an exemplar car at the crash location showed that Traffic Aware Cruise Control could be engaged but that Autosteer was not available on that part of the road.

As the footnote indicates, Traffic Aware Cruise Control handles acceleration and deceleration, while Autosteer keeps you in your lane. Autosteer requires lane markers to be utilized, which that stretch of road did not have. My takeaway from this is that the guys in the car would not have been able to engage the Autopilot, as Tesla has claimed. The investigation is ongoing, and as noted the NTSB has not yet determined a cause, so we still don’t really know what happened. I remain curious about this and look forward to the final report.

UPDATE: Here’s a longer version of the story.

Tesla disputes official account of that autopilot crash

Can’t say I’m surprised.

Tesla officials are disputing numerous assertions of local police related to a recent deadly crash in the Spring area, reiterating statements by founder Elon Musk that autopilot was not engaged on the vehicle.

Further, company officials contradicted officials in Houston, saying the car did not have a driver during the crash.

“The steering wheel was indeed deformed, leading to the likelihood that someone was in the driver’s seat at the time of the crash,” said Lars Moravy, vice president of vehicle engineering for Tesla.

The comments, the company’s first statements on the incident aside from Musk’s tweets, came during a Monday evening conference call related to the company’s first quarter financial performance.

The crash, which has drawn national attention because it involved a Tesla, is under investigation by Harris County Precinct 4 as well as the National Transportation Safety Board and National Highway Traffic Safety Administration.

[…]

Moravy said a review of the car’s data and the conditions on the street – which lacked painted lines – make use of autopilot impossible.

Further, he said the vehicle’s cruise control only accelerated to 30 mph because of the street’s speed limit as set in Tesla’s traffic computers, and that when the driver unbuckled it slowed.

All of the seat belts in the car were unbuckled at the time of the crash, Moravy said.

Reached Monday evening, Herman said local police were aware of Tesla’s claims but would not comment on them.

“Our case is still under investigation,” he said.

See here for the background. I mean, maybe Tesla is correct, but given their overall track record with the NTSB and the clear demonstrations of how to fool the autopilot system into thinking there’s someone in the driver’s seat, I’m not inclined to take their word for it. Let’s have a full investigation and see where that takes us. The Washington Post has more.

What’s up with that Tesla autopilot crash?

I assume you’ve heard about this.

Woodlands Fire department, Montgomery County Hospital District and Cypress Creek EMS were dispatched around 9 p.m. Saturday to a fire in the woods in the Carlton Woods Subdivision on Hammock Dunes Place.

Several neighbors had called reporting a fire in the woods, and that a car had crashed and exploded, Palmer Buck, the Woodlands Fire Department chief said.

When the responding units arrived at the scene firefighters discovered the bodies of two males in the 2019 Tesla Model S, according to the Montgomery County Police Reporter. One male was in the front passenger seat and the other in the rear passenger seat.

Harris County Precinct 4 Constable Mark Herman told the Associated Press on Monday that investigators are “100% sure” that no one was driving the car.

Federal investigators are on the scene to find out what happened. The claim that no one was driving the car at the time is of course of interest, for all the obvious reasons. Elon Musk has publicly disputed this assertion, claiming that it’s not possible because the Tesla’s autopilot function will shut down if no one is in the driver’s seat. It turns out that’s not exactly true.

Consumer Reports engineers easily tricked our Tesla Model Y this week so that it could drive on Autopilot, the automaker’s driver assistance feature, without anyone in the driver’s seat—a scenario that would present extreme danger if it were repeated on public roads. Over several trips across our half-mile closed test track, our Model Y automatically steered along painted lane lines, but the system did not send out a warning or indicate in any way that the driver’s seat was empty.

“In our evaluation, the system not only failed to make sure the driver was paying attention, but it also couldn’t tell if there was a driver there at all,” says Jake Fisher, CR’s senior director of auto testing, who conducted the experiment. “Tesla is falling behind other automakers like GM and Ford that, on models with advanced driver assist systems, use technology to make sure the driver is looking at the road.”

Our demonstration comes as federal and local investigators continue to probe the cause of a fatal crash Saturday in Texas in which an apparently driverless 2019 Tesla Model S struck a tree, killing the vehicle’s two occupants. Harris County Precinct 4 Constable Mark Herman, who was on scene at the crash, told CR that he’s almost certain that no one was in the driver’s seat when the vehicle crashed. (The Model S in the crash and our Model Y are different models, but they both have Autopilot.)

We tried to reach Tesla to ask about the Texas crash but did not hear back. Tesla CEO Elon Musk tweeted Monday evening that data logs recovered from the crashed Model S “so far show Autopilot was not enabled,” and he suggested that it would not be possible to activate Autopilot on the road where the crash took place because of the lack of painted lane lines. The National Highway Traffic Safety Administration and the National Transportation Safety Board are investigating the crash, which occurred on a winding road in Spring, Texas, outside of Houston.

CR wanted to see whether we could prompt our own Tesla to drive down the road without anyone in the driver’s seat. So Fisher and Kelly Funkhouser, CR’s program manager for vehicle interface testing, took our 2020 Tesla Model Y out on our test track. Funkhouser sat in the rear seat, and Fisher sat in the driver seat on top of a buckled seat belt. (Autopilot will disengage if the driver’s seat belt is unbuckled while the vehicle is in motion.)

Fisher engaged Autopilot while the car was in motion on the track, then set the speed dial (on the right spoke of the steering wheel) to 0, which brought the car to a complete stop. Fisher next placed a small, weighted chain on the steering wheel, to simulate the weight of a driver’s hand, and slid over into the front passenger seat without opening any of the vehicle’s doors, because that would disengage Autopilot. Using the same steering wheel dial, which controls multiple functions in addition to Autopilot’s speed, Fisher reached over and was able to accelerate the vehicle from a full stop. He stopped the vehicle by dialing the speed back down to zero.

“The car drove up and down the half-mile lane of our track, repeatedly, never noting that no one was in the driver’s seat, never noting that there was no one touching the steering wheel, never noting there was no weight on the seat,” Fisher says. “It was a bit frightening when we realized how easy it was to defeat the safeguards, which we proved were clearly insufficient.”

There’s video at the link, and I also recommend listening to Friday’s What Next TBD podcast, which discusses this crash, Tesla’s spotty record with its autopilot feature, AI and the driverless car question, and more. Tesla is not currently cooperating with NTSB on this, which has drawn some ire from Rep. Kevin Brady, who represents The Woodlands. I probably won’t follow this obsessively, but as driverless cars are an interest of mine I will keep an eye on it.

Speed kills

Good long read from the Chron about our dangerous roads and highways. There’s too much to cover here, so I just want to focus on the why we all speed so much.

Houston drivers likely speed, at least in part, because they believe no one with authority is paying attention.

A Chronicle analysis of municipal court data shows that Houston-area law enforcement’s largest agencies are deploying fewer officers for road enforcement and ticketing fewer drivers, even as fatalities increased in the past two years and the area grows in population.

Houston police officers ticketed 41 percent fewer drivers in 2017 than they did in 2012, even as the number of vehicle miles traveled in Houston grew 23 percent.

That reflects a national trend of less traffic enforcement, according to Hersman, the former chairwoman of the NTSB. Federal statistics show that the share of people coming into contact with police through a traffic stop dropped about 11 percentage points from 2002 to 2011.

“We certainly understand what law enforcement is being asked to do and what they deal with, but the reality is fatalities are going up on our roadways,” Hersman said. “What we are seeing nationwide is law enforcement is not doing traffic enforcement.”

Harris County sheriff’s deputies, for example, issued 28 percent fewer speeding tickets in 2017 than they did in 2015, even though the county gained 100,000 people during that period. Houston police officers issued 16 percent fewer speeding tickets in 2017 than in 2015. Texas Department of Public Safety Troopers operating within the greater Houston region are the exception; they issued 11 percent more tickets for speeding than they did in 2015.

I mostly travel on I-10 these days, and I do see (usually unmarked) patrol cars on the shoulders, and occasionally a pulled-over vehicle getting cited. But this is the exception, and there’s nothing quite like the joy of being tailgated when you’re already doing over 70 on a road with a speed limit of 60. I don’t have any solutions to offer here – we could reduce speeding and the mayhem that accompanies it with higher levels of patrol, but of course that’s going to require more patrol officers, and that’s not in the cards. I just miss working in a part of town where I didn’t have to take highways to get to the office.