Tesla Model S Crashes Into Back Of Fire Truck, Driver Blames Autopilot

Tesla Autopilot


Autopilot crash

Tesla Model S crashes into parked fire truck … driver blames Autopilot (Image Credit: Culver City Firefighters @CC_Firefighters via Twitter)

Was it the fault of Tesla Autopilot or the fault of the driver?

As autonomous vehicles begin to take public roadways, this is a question that will come up often. There’s only one right answer …

It’s always the fault of the driver, at least until the technology improves and regulators and insurance companies come up with new policies.

Tesla Autopilot

Tesla Autopilot was investigated as part of a fatal accident last year. Again, the driver was not properly engaged. The technology was not found at fault, and actually proven to reduce crashes if used correctly (via ABC News/Bobby Vankavelaar)

We’ve seen time and time again that drivers blame crashes on the car. This was true even before self-driving technology.

“Officer, the gas pedal stuck … the brakes stopped working … the engine stalled.”

Fortunately, no one was hurt in this recent incident.

Let’s get a bit more serious for a moment. Yes, if Tesla is marketing vehicles that make people think it’s safe to drive without paying attention or give them a false sense of reassurance, the results could be devastating. While no one was injured here, someone easily could have been.

However, the automaker has made it abundantly clear that the driver must remain engaged and alert. The car repeatedly reminds you to keep your hands on the wheel and stay in control of the vehicle. It chimes and gives visual alerts like “Hands Required Not Detected.”

We recently reported about a drunk driver trying to use Tesla Autopilot to get away with his crime. Again, no one was injured, and thankfully, the Tesla was most likely the reason he’s safe and no one else was hurt. But, it’s surely not a viable excuse.

Reports from the recent fire truck incident conclude that Engine 42 was on the freeway taking care of an incident when the Tesla Model S slammed into the back of it going 65 mph. The incident occurred on the 405 Freeway near Washington Boulevard in Los Angeles (Culver City).

The driver told officials that the car was in Autopilot mode, but there is currently no proof of such. Anyone traveling 65 mph on a freeway, Autopilot or not, should take notice of a huge red truck that’s parked and not drive into it without avoiding and/or braking.

The moral of the story …

As we ride into the advent of vehicle autonomy, please keep your hands on (or very close to the wheel), be prepared to take control and brake, and keep your eyes on the road.

Source: CBS Los Angeles

Categories: Crashed EVs, Tesla

Tags: ,

Leave a Reply

79 Comments on "Tesla Model S Crashes Into Back Of Fire Truck, Driver Blames Autopilot"

newest oldest most voted
Another Euro point of view

I don’t understand at all this autopilot thing.

I mean if situation is that even with the autopilot on I need to sustain the same amount of attention than if autopilot is off, isn’t it better not to switch on the autopilot at all ?

I mean if an autopilot is not automatically/sufficiently braking for an obstacle isn’t it better to just drive the car, that being the best way to ensure one’s one’s keeps attention to the road ?

Could someone explain in case I’ve got something wrong ?

If, if, if…
The thing is we don’t know for sure it was on autopilot. IF it was, i would expect it to stop. A big red truck in the middle of the road is impossible to miss.

The first rule of Tesla Motors Club is you blame autopilot.

The second rule of Tesla Motors Club is you blame autopilot.

—Tyler Dryden


“It’s Always The Fault of the Driver” Unless…The Car is Void Of a Steering Wheel, Brake & accelerator Pedals.. THAT”S ALL!!!!



I guess ‘Key Fob’ is not equal to ‘key’ for some Tesla Drivers!
“Sorry Officer, FOB was driving, not me!”

As I understand it, all automatic braking functionality in cars (not just Tesla) will not react to a stationary object/vehicle. I believe it requires the thing to be moving (although much slower). I’m guessing this eliminates the risk of it accidentally picking up something that isn’t moving which is not a vehicle (e.g. a bollard, barrier, sign etc).

Apparently not – check out the various YourTube videos featuring this question.

(⌐■_■) Trollnonymous

” isn’t it better not to switch on the autopilot at all”

Exactly. Why even bother IMHO.

If they had an option to not have AP at all on their cars, I would purchase that one.

Below is what happens when Teslas are driven with Autopilot off. Tesla drivers, crashing into all sorts of buildings and structures since 2012. Here’s a crash from just last week.


This has been reported before. I don’t understand why Tesla allows the car to be driven with the Falcon Wing doors open. In my opinion the car should refuse to move at least without specific override from the driver.

IIRC, the Model X can be driven up to about 30 mph with the falcon-wing doors open. At that speed, the aerodynamic drag on the open FW doors would probably misalign those balky, troublesome FW doors. Why Tesla allows these speeds with the FW doors open is beyond me.

I’ve also read of the FW doors opening by themselves for no apparent reason while driving at high speeds. I’m not sure if this would cause the Midel X to slow down or come to a stop.

I agree that a Model X shouldn’t move with a FW door open, at least not without an override from the driver. Seems odd that Tesla doesn’t add that function to the software.

As far as FW doors spontaneously popping open while driving down the road… well, shockingly enough, some people post things to the internet which are not entirely, completely, 100% true.

Especially when it comes to Tesla, which sadly has attracted so many serial bashers and FUDsters… some of whom have already posted comments to this very discussion thread.

(⌐■_■) Trollnonymous

There are plenty of cars that you can drive up to freeway speeds with the doors open.

Yes, but they are hinged at the front, not nearly as likely to be damaged or cause damage. Also normal car doors don’t open automatically just because you bumped your key fob in your pocket.

I think you’d have a hard time keeping a front hinged door open while traveling at 65 mph on the freeway. That being said, there really shouldn’t be a restriction on moving a car based on the door being open, simply because you wouldn’t want to be stranded simply because the door sensor went bad.

One time my truck’s door got damaged in a freak accident (parking brake failed to hold due to excessive weight in the bed and the door got wedged against a parking post). I ended up having to strap the door shut. But I’d have been really pissed if had to call for a tow simply because the vehicle refused to move with the door “open”.

There’s a reason rear hinged doors are called “suicide” doors… suck you out with them if you crack ’em even a little.

AP is an option. The hardware is there but not usable if you don’t purchase the software, so you can easily buy a Tesla without AP enabled.
I see you on these forums all the time, surely you understand this is the case?

(⌐■_■) Trollnonymous

I prefer not to have the hardware either.
I don’t care if it saves $30 or $150 without the hardware, I just don’t care for it.

Driving requires your full attention and hands on the wheel.
AP requires the same thing.
WTF is the purpose?

The greatest luxury of Autopilot is in stop and go traffic. This is about the only place people can take their eyes off the road while stopped, and let the car’s acceleration raise their attention again. Of course, it’s also nice to have a free hand.

The fire truck was likely stopped in the lane. Autopilot doesn’t see stopped cars as well as moving ones, even if humans can. Autopilot doesn’t adjust well for stationary off-set intrusions, either (youtubes). So, even if one risks not having their hands on the wheel, it is that much worse to take eyes off the road when you’re going this fast (pretty obvious from the pic).

Humans are safer than computers, if we’re talking about full control. This data point makes me think of what that pilot, “Sully”, was onto. His schtick is having pilot-and-computer understand each others limits.

People love the option (me included). Otherwise, Tesla wouldn’t have stamped “$5,000” on it for their $35,000 car.

It’s the same principle as regular cruise control: it reduces the physical and mental effort involved in driving.

Except when it doesn’t.

(⌐■_■) Trollnonymous

Cruise control requires you to have your hands on the wheel and alert at all times.
I think that’s in every manual too.

if situation is that even with the autopilot on I need to sustain the same amount of attention than if autopilot is off, isn’t it better not to switch on the autopilot at all ?

Tesla Autopilot + AutoSteer doesn’t try to text while driving.

It doesn’t get drunk and try to drive.

It doesn’t fall asleep at the wheel.

It doesn’t try to read or eat or put on makeup while driving.

It’ doesn’t get so involved in talking to a passenger that it forgets to watch where it’s going.

It doesn’t suffer from “highway hypnosis”.

It doesn’t get so angry at being cut off by another car that it gets into a road rage incident.

All this and more, is why The NHTSA says that Tesla cars equipped with Autopilot + AutoSteer have a 40% lower accident rate than Tesla cars without them. (source linked below)

That ought to be the end of the discussion. Sadly, it won’t be.

“The thing to keep in mind is that self-driving cars don’t have to be perfect to change the world. They just have to be better than human beings.” — Deepak Ahuja, CFO of Tesla Inc.


“The NHTSA says that Tesla cars equipped with Autopilot + AutoSteer have a 40% lower accident rate than Tesla cars without them.”

Maybe it is enough to repeat the same debunked lie over and over again? It isn’t exactly what NHTSA reports says. Stop lying Pu-pu!

It is not about some “Autosteer”, but about AEB (automatic emergency braking) that was introduced with Autopilot package, thanks to MobilEye. The same 40% airbag deployment reduction rate is for other automaker AEB as well.

Wow, zzzzzzzzz accusing Pushy of lying? That’s not the pot calling the kettle black; it’s the pot calling the white porcelain bowl black!

You ought to change your screen name to “Mr. Hypocrite”, troll.

Here’s the actual graph that’s in the article I linked to, since you can’t be bothered to look at the actual evidence, and prefer to keep posting your Tesla Hater cultist lies:


Another Euro point of view:
“I mean if situation is that even with the autopilot on I need to sustain the same amount of attention than if autopilot is off, isn’t it better not to switch on the autopilot at all ?”

Yes, it obviously safer to drive yourself instead of paying attention to a drunk driver and being ready to take over the wheel in split second when it swerves off the road, giving away control 0.001 seconds before impact and claiming “I’m not responsible anymore!”

Still, it this case “Autopilot” looks as classic distraction technique for Musk disciples. It is more likely about failed auto-braking, not active driver assistance. Auto-braking should function by default with or without driver assistance, and is standard equipment even in cheap toyotas now. Was the truck too red or sky too white this time?

“It is more likely about failed auto-braking, not active driver assistance.”

Does any auto maker’s ABS (Automatic Braking System) work with stationary objects, as was the case here with the stopped fire truck?

I’m not aware that any of them do; would be nice to know if other auto makers have a better system.

To answer my own question:

Yes, ABS has been shown to stop a car from running into the proverbial brick wall. But probably not every time.

(Unfortunately my Google-fu has failed to find the photos or video I recall seeing of an ABS “crash test” without a crash, showing a car stopping just short of hitting a concrete barrier. 🙁 )

If the automaker sells a device called “Autopilot” the attention of the driver automatically decreases when the “Autopilot” is engaged.
In China Tesla is no longer allowed to call it a “Autopilot” simply because it ain’t…

Having had the (wonderful) benefit of many thousands of miles using Ap in my Model S, I can tell you that the most obvious benefit is that long drives are much less tiring than if done manually. My theory as to why is that although an experienced driver does not notice that, as they drive, they are using lots of muscles all over the body with great regularity even if they are not moving much. On top of that, in order to drive competently, the brain and nervous system is actually working very hard: maintaining a lane/constant speed/safety gap and all the while looking out for potentially dangerous situations developing, etc, etc. If you don’t believe me, just consider how hard you had to concentrate on not crashing when you were a very novice driver – or watch one being given a lesson. Concentration is absolute for newbie drivers as there is so much to think about. With experience, most of this goes to the back of your mind. Experienced drivers can still see this when trying to negotiate a complex junction whilst trying to hold an intelligent conversation with a passenger – and this is one reason why driving… Read more »

Yeah coast to coast at the end of 2017

“. . . the Tesla Model S slammed into the back of it going 65 mph.”

Give me a break. I call BS. The Tesla was NOT going anywhere close to 65 mph when it slammed into the fire truck. The pic above shows a low-speed collision. Either the Tessa wasn’t driving very fast before the impact or automatic emergency braking slowed the car down dramatically from 65 mph. Either way, that is not a 65 mph impact, because the damage would have been much, much worse at 65 mph. It’s not even close.

Well the Model S did get 5.3 stars on the NTHSA crash tests, more than any car ever.

The absence of the big ICE engine slamming in your face and exploding the cockpit in a front collision is a unique EV safety feature.

Tesla took advantage of this by putting aluminum struts just at the right places to reduce the impact shock.

For the past couple of decades, all modern ICE car have been designed to push the engine and transmission under the car in a frontal impact crash and not into the passenger compartment.

That one looks more like higher truck with no proper under barrier at the end like in ancient times. It is amazing American trucks still don’t have these at the sides even today.

Tesla will publish the drive logs shortly which throw the owner under the bus and show in detail how he was either abusing AP or didn’t have it activated. And absolve itself of any fault.

As they should! When GM goes online with their AP system (hopefully very soon) wouldn’t you want the fakers be exposed?

Supercruise is already available to my knowledge.

Is this the same as adaptive cruise? or is it something more?

I agree, just look at the 40mph full frontal crash test videos and then remember that impact varies as the square of velocity. That means this crash at 65 mph had 2.64 times the impact of the tested vehicle.

The Tesla collided with a stationary fire truck, the equivalent of a brick wall, going full highway speed and everyone’s ok. Is that not the real story here?

(⌐■_■) Trollnonymous

C’mon man, stop using logic……lol

Maybe! Or, the real story is “Dummies come in any and every financial status!” Or: “Another Tesla Battery Pack is soon to come on the used Market!”

I think the real story is “Yet another driver tries to blame Autopilot for his failure to stay awake, not drunk, and watching where he was going when he was supposedly driving the car.”

So far as I know, Autopilot isn’t yet intended to stop a car when a stationary obstacle is detected, not even one in the vehicle’s lane. Something about “too many false positives”, hmmmm?

It is not equivalent to brick wall, the truck looks clearly moved from the impact, so deceleration distance was much longer. It is not loaded semi, just a firefighter truck, and Model S is relatively heavy. Who knows what the speed was without investigation.

“…the truck looks clearly moved from the impact…”

How so? I see no evidence of that whatsoever. For example, there are no skid marks behind the fire truck’s tires.

(⌐■_■) Trollnonymous

Drivers fault 100%.
I bet AP wasn’t even on.

~IF~ it was really at 65mph, the Model S would be much more under the the fire truck upto/into the cabin.


Here is 100 km/h (62 mph) crash test.

Passenger compartment is more or less in place like in this Tesla. Model S may also have a bit longer crumple zone, and the truck moved on impact.

I don’t think his “neural net” was on. Or it was busy looking for rain. 🙂

HAHA! Seriously though, I thought “S’s” only had trouble with totally white vehicles.

There are some white stripes here but the majority of it is Red.

When Autopilot claims to be SAE Level 4 or 5, then you can blame Autopilot for crashes (unless the other driver is also human).

Until then, it’s always human pilot error.

It is extremely unfair to expect a person to be fully alert when he / she is doing essentially nothing except watching the road. That is why locomotive drivers have a “vigilance device” they have to actuate every so many minutes.

What year Tesla? Automatic emergency braking should minimize or prevent such collisions regardless of Autopilot status. And maybe it did minimize the collision in this case (it doesn’t look like a 65 MPH collision).

All new cars in the US will have AEB by 2022, so this isn’t just about Tesla. Obviously, there are tradeoffs, but I hope those systems can be designed and configured to safely handle the vast majority of common scenarios, and not just a limited range of speeds, etc.

I dislike the high reliance Tesla puts on radar detection for braking. Radar works by metallic reflective surface back to the car. The firetruck was at an angle, so all radar beams would be reflected at angles away from the car. A better camera based visual system would have detected the truck and initiated the brakes earlier.

Autopilot AI: “So, THAT’s what the back of a Fire Truck Looks Like, coming up fast! OK, it seems that is an Obstacle! ‘Note to Mothership: Avoid these items!'”

The last thought of the car was “I wonder if it will be friendly?”


“A better camera based visual system…”

Cameras are “better” than radar for detecting obstacles in the same way that a Yugo is “better” than a Rolls-Royce.

I have rarely read any comment posted to InsideEVs which was as flat wrong as that one.

I agree with most of your posts, but obviously not this one. We see and drive with our eyes, not radar, and are able to detect almost any object in our path. If radar was so good, why didn’t it stop this car in time? Why do you think cameras wouldn’t see objects?

There is evidence in the photo above that the Tesla brake was fully engaged before impact….by whom is anyone guess. The fire truck has a low profile and in order for the T to get under it the way it did the brakes need to be applied hard in order to make the front end dip. My guess is the autopilot was not engaged. This reminds me of the rv driver lawsuit when he placed the rv in cruise control to go to the bathroom.

…except that the “RV driver used cruise control to go to the bathroom” is just an urban myth. And your assertion that the Model S couldn’t possibly wind up with its nose underneath the rear of the fire truck unless it was heavily braking, may be equally incorrect. The impact itself may well have shoved the nose of the Model S underneath.

However, I do agree with the various comments upstream which say this doesn’t appear to be a 65 MPH impact, and that there was likely some slowing (quite possibly some braking) before impact.


You are right, either the driver or the AP did apply the brakes, just too late. I think when the car got close enough the AP did work, even if the driver had already applied the brakes.

You want to see driver assist technology done right? Go drive a new Subaru Outback.

This somewhat puts thing in perspective with regard to GM and others not only using camera’s but lidar to physically gauge objects vs a camera which could get fooled under certain conditions. I would think auto braking would have been engaged. Interested to hear Tesla’s interpretation.

A LiDAR-enabled SLAM* system is likely what cars will require before they can reliably detect obstacles stopped in a traffic lane, as with the fire truck here.

But I’m pretty sure nobody has yet started selling cars equipped with a LiDAR-enabled SLAM system. So far as I know, those are found only on self-driving testing fleets and prototypes.

*SLAM stands for Simultaneous Localization And Mapping technology, a process whereby a robot or a device can create a map of its surroundings, and orient itself properly within this map in real time.

Is there a quote of the driver blaming Autopilot or is this just another misleading headline?

It’s in the attached Tweet from the Culver City fire department.

When “failed AP” allegations are made, Tesla usually and quickly provides monitoring data to refute the claims. I’m waiting for the Tesla response…

Cannot imagine crashing in in other vehicle at 65 MPH and just ‘getting out’…

Also even IF he had Auto Pilot on, what prevented HIM from stopping the car before it hit the HUGE red thing in the road?

I think it’s pretty clear that this is an attempt by the driver to say “Don’t blame me, blame Tesla Autopilot!” From his position, it doesn’t even matter whether or not it was really engaged; he’s just trying to shift blame.

Perhaps he’s not aware that by law, he’s just as responsible one way as the other. Or maybe he’s trying to shift blame merely to reduce his personal embarrassment.

So, even without Autopilot, doesn’t Tesla comes with AEBs that would automatically brake unless the driver attempted to brake on his own?

That’s what i was thinking too but then again, on most cars that have this you can deactivate it.

From various reports I’ve seen, I don’t think ABS is all that reliable. And that applies to all auto makers’ ABS systems, not just Tesla’s.

Speaking from a computer programmer’s viewpoint, I think the problem is all the false positives. If an ABS system stopped the car every time it detected any obstacle near the traffic lane, your car would be stopping so often that you’d soon shut it off. If the sensitivity is, so to speak, “turned down” sufficiently to eliminate nearly all the false positives, then — apparently — it sometimes misses real obstacles, as perhaps was the case here.

But there are other possibilities. Perhaps the driver was lightly pressing the brake pedal, which (if my understanding is correct) would disengage ABS. If the driver was rubbernecking at whatever accident or situation the fire truck was involved with, instead of watching where he was going, then he might well have been slowing down by light braking.

Tesla driver had autopilot set to 65 mph. Since there was a accident ahead, it was probably stop and go.

The car that autopilot was following moved over to the right. Autopilot accelerate without recognizing the stop firetruck and slammed into it.

Whether flying a plane with autopilot or a car with autopilot, you must pay attention. Planes still crash even with 2 pilots and an autopilot. I have a model X. Right now autopilot is a fancy cruise control. You can still crash with autopilot on. People want to blame inattentiveness on their part. Driving a car is dangerous. They were lucky they were not killed thanks to Tesla engineering. Just like a plane or car, you are captain of the vehicle and take full responsibility. I would love to blame someone else vs the person that is directly responsible! I love it when the news people check to see if the autopilot was on or off. What about the other manufacturers that kill people in their cars

Auto pilot or not,Automatic emergency braking didn’t work properly.Tesla should stop concentrating on its circus acts and start making proper safety tech.