Featured Safest SUV Ever Tested by NHTSA

I suspect the Insurance Institute for Highway Safety (IIHS) is already planning some new test to try and fail the car.
Tesla's electric car falls short in new IIHS crash tests

Bob Wilson

 

Ouch! All Tesla gave was excuses.

 

I'm sure they poured over that data and are working on it.
After all, the Prius also failed the small overlap test the first year.

I would like to see the X tested by the IIHS. But the rate of tests is, understandably, slow.

 

I was and remain unhappy with the IIHS since there have been at least two cases where the Prius was top rated only to get down rated within six months by the IIHS. Ok, so safety standards change but IIHS went out of their way to call out the Prius in their subsequent press releases to nearly the exclusion of Detroit products that also failed. It became a regular pattern and I have no doubt that Tesla is in their sights.

As for USA Today, it is so low on my threshold, I hadn't even considered it anything but Rah! Rah! reprints of press releases free from critical review.

Bob Wilson

 

I generally find IIHS test results more useful that the basic tests run by the government, but that is just my opinion.
Tesla rated high on the easier government tests but failed the more rigorous IIHS tests.
I read the article but I generally consider the government tests set so low a bar they generally can be ignored when passed.

The tests were for Tesla models built after October 2016, according to the IIHS site.

 

IIHS wants to incentivize car makers to create safer cars so insurance companies can save money. After all, who is their funding source. They aren't out to get your favorite manufacturer. And once there are a significant number of models passing the current tests, why not announce some future tests that statistics have proved to be relevant based on statistics from crashes so that future models can be designed to better protect the occupants so the death benefits don't have to be paid.

Better, not perfect. And yes it is hard to overcome the laws of physics.

 

"To qualify for TSP+, the IIHS testing parameters require each vehicle to earn “good” ratings in five crashworthiness tests – small overlap front, moderate overlap front, side impact, roof strength and head restraints – as well as an “advanced” or “superior” rating for front crash prevention and an “acceptable” or “good” rating in headlight testing." (from a cleanmpg.com article)

For which test would you like your car to score less than "good"?

 

Does it really make people safer, or does it make higher premium/repair cost. IIHS has been a major lobbiest for larger, more expensive vehicles. The have effectively lobbied cafe rules to favor, you guessed it, trucks and SUVs. Are we really safer with these bigger vehicles? Statistics do not show it. They do show that if you plow into a smaller vehicle with your bigger one, the people in the bigger vehicle are less likely to be hurt.

So what about this small offset crash thingy that was telling us that my 2010 prius and the 2016 model S is somehow unsafe?

We can look at the studies. Roughly 6.4% of frontal accidents according to the one I found, are small offset. When the accident is 2 vehicles the risks aren't really great of serious injury, but if you hit a tree ... well it could be serious. NIH study of all the data said it is at worst a moderate risk in the scheme of things.

Driving a car is dangerous. You may cross the highway and it a semi truck. You could accelerate into a tree.

On the other hand the number of prius injuries in gen III from small offset before toyota rejiggered the car to pass the test is extremely small. The model S, I don't think 1 major injury has occurred from a small offset test, that would be 0% so far. Could it happen? Sure, as it could in a vehicle that passed the test.

Don't get too excited when a scientific sounding organization, that really is a group of government lobbyists tells you, that you need to buy something new, because that thing you have been using safe for years now is somehow unsafe. Really I'd love my car to have an electronic nanny to stop me from hitting a tree if somehow I had a heart attack or something and couldn't drive. I think i am careful enough not to fall asleap and hit something. I don't really need titanium shields around me for the next IIHS test. Is your car older than 15 years old? Chances are you can buy a safer one. I doubt that many cars are safer than the model S, no matter what the IIHS says. There have been at least 4 crashes in the S that most IIHS top safety cars would have had major injuries, but the model S occupants were fine. The thing is built to survive.

 

Just fyi, here is the list of safest cars. These are the safest cars on the road - May. 25, 2017 . Tesla model X hasn't been around long enough to be considered ;-)

Of note, the Audi A6 and Mazda CX9 failed the new test, but somehow are in the top 11 safest cars. ;-)

If testing made perfect then frontal crash accidents would have been decreased 100%, unfortunately there are many reasons that this is 61%, still a marked improvement, but no where close to perfect. Seat belts, Crumple zones, air bags can't prevent them all. Brakes and active systems help more as does removing impaired and distracted drivers. All these safety devices probably mean drivers feel free to drive less safe. That small offset crash is most dangerous not in collision with another vehicle, but with a fixed object like a tree or a pole.

The model S was stopped not because of small offset, but by major driver error. The first 2 fatalities, the drivers drove off cliffs. Tesla has since added electronic nannies to automatically steer in emergencies to prevent this. The third a car thief hit a post going 100 mph. A 30 mph test is not going to stop that kind of abuse but software that the driver can't disable might.

 

Note the absence of any rear impact crash test.

 

Since we discussing Tesla safety, a detailed report is available:
https://www.ntsb.gov/news/press-releases/Pages/PR20170619.aspx

There are many summary reports about this accident but I tend tor read the originals when I can. NOTE: Accident reports are not for the faint of heart.

Bob Wilson

 

Driver killed in Tesla self-driving car crash ignored warnings, NTSB reports

Always good to read the originals, if you think reporters may be leaving out key details. Bottom line, according to the driver's family lawyer, the tesla data shows that the driver was not watching a movie. According to the NTSB the driver, not the car was clearly at fault, as the driver ignored many warnings and had lots of time to avoid this accident even though he was speeding. Witnesses also seemed to indicate, that like in many accidents, the truck driver could have also avoided it, but did not. Two different drivers were not paying enough attention.

The tesla system, is a wonderful way to build a safety system. It learns from many drivers things about unsafe driving conditions, and programers adapt the hardware for future models, and software for all models to help avoid these accidents. It is necessary that drivers pay attention and correct the system when it makes mistakes. It already makes these cars safer than most. Tesla already has corrected to account for this situation in the future. There will be other mistakes. The driver had his hands on the wheel, and probably was paying attention, a very low percentage of the time. If he had been momentarily distracted, the system would have kept him safe. He appears to have been paying little attention at all. The systems are not good enough for that. Google's self driving cars only go up to 25 mph, and have much better and more expensive sensors. The tesla was going 74 mph when the accident happened, and the driver clearly was not paying attention for the 7 seconds that were critical in this accident.

 

Admittedly, I haven't read the report, but here's a question that came to mind: The going understanding is that the cameras could not distinguish the color of the side of the truck trailer from the color of the background sky. It seems like radar should have had no problem detecting the truck, though. The best guess I've heard is that the radar beamed under the truck, and just didn't give enough of a return to trigger the system to emergency-stop the car. Either that or the truck trailer was at an angle that reflected almost all of the waves to the left, and, again almost no return.


iPhone ? Pro

 

The version of software being run back then has been changed to account for bad driver behavior.

1) If the driver doesn't heed warnings to pay attention, autopilot will disengage. Driver is forced to at least periodically put a hand on the wheel.

2) Auto braking needed autopilot to be on and the vision system to see something. Now if the radar system sees a crash about to happen, it will brake whether autopilot is on or the camera sees anything.

Number 1 alone would have probably saved the driver. Number 2, we don't know, but the car would have been going slower when it hit. The driver of the truck is being charged with failure to yield the right away, and refused to talk to the ntsb. When 2 drivers do the wrong thing, its hard for new software to know all situations. Still tesla has tried to prevent this particular situation in the future.

NTSB estimated that the autopilot has prevented 40% of accidents. That means drivers need to still help on the other 60%. We already know many cases where change number 2, braking under radar, even if the driver hasn't enabled it, has saved lives. As Hill says, tesla has improved hardware for new cars too. Still software improvements is helping save lives even on older hardware.