Tesla Vision earns IIHS "Superior" rating avoiding crash in *all* pedestrian tests

2022 Model Y received TOP SAFETY PICK+ today with both front crash prevention tests (vehicles and pedestrians) getting Superior rating. Tesla has consistently gotten Superior rating for vehicle-to-vehicle, but even with a Tesla Vision re-test with updated software back in June, the vehicle-to-pedestrian score "only" scored Advanced.

However with the latest Tesla Vision tested in the Model Y and re-tested with Model 3, not only did IIHS rate it Superior, all pedestrian tests avoided collisions including traveling at 25mph with a small dummy running out from behind parked vehicles obstructing view. Previously with radar Autopilot, the crossing child test only reduced the speed by 5mph.

What other safety systems can completely avoid collisions in all IIHS tests? And of those, how many vehicles have it as a standard feature? I wonder how long until even Tesla vehicles with radar get switched over to Tesla Vision to improve safety.

 

[sleepydoc]

It’s an impossible comparison to make. Pedestrian fatalities are reported and tracked. False AEB activations are not. Additionally, false activations likely cause more damage than fatalities. If I’m driving on I94 through Minneapolis in rush hour and my car suddenly slams on the brakes I may get rear ended but it won’t kill me. Or the car behind me may stop but the car 2 or 3 cars back doesn’t stop and rear-ends someone else, causing an accident (or several) much further back. Now add the soft costs of increased congestion, etc from these activations.

 

[cobra]

It’s an impossible comparison to make. Pedestrian fatalities are reported and tracked. False AEB activations are not. Additionally, false activations likely cause more damage than fatalities. If I’m driving on I94 through Minneapolis in rush hour and my car suddenly slams on the brakes I may get rear ended but it won’t kill me. Or the car behind me may stop but the car 2 or 3 cars back doesn’t stop and rear-ends someone else, causing an accident (or several) much further back. Now add the soft costs of increased congestion, etc from these activations.

Maybe. Or maybe this is Tesla's way to finally break tailgaters, one car at a time!

 

[username123]

Thanks. I wonder if both have relatively high phantom braking too as others pointed out achieving high true positive often comes with higher false positives. Similarly, some people say they prefer no false positives, but that likely comes with a human cost in terms of safety. Hopefully Tesla can significantly improve precision while maintaining or even improving recall with FSD Beta stack then deploy it to the whole fleet even with standard Autopilot.

I have a 2020 Mazda 3 and only had phantom braking once when reversing into my driveway which has a slight incline.

It saved me a couple of times from backing into low objects at night time and applied the brakes harder than I was when I had a car in front of me slam their brakes.

I’ve never had any issue with the radar cruise control and personally find it better than the Tesla’s. When cars are braking in front, it’ll gradually slow down and try its best to never completely stop by creeping forward slowly

 

[Two-rocks]

A student at my daughters college was hit by a car and killed, at night, in the rain - no foul play/alcohol - just walking.

When the weather is bad on my boat I use RADAR, no other way.
When the evening arrives, otherwise clear, my tesla brakes for everything, and nothing.

The clock is ticking on the warranty expiring and gaining root access, someday the car might actually be mine.

 

[username123]

If by 'it' you mean a superior rating, then yes.
If you mean Tesla level, then no. At 37 mph the Mazda issued a warning 1.8 seconds before a collision would have occurred. Tesla vision warned 2.7 seconds before collision.

This difference is huge -- just imagine streets with poor traction.

Yes but for Tesla Vision in Parallel adult test they provide a warning 2.8 before impact

You’re comparing a $23k car (base model in Canada) that has had this equipment as standard since 2019 and has always avoided all the crashes with no widespread reports of phantom braking to a $60k car made by a tech company that was just now able to avoid impact after collecting data from their fleet for years with thousands of people complaining about phantom braking.

Moreover, why does it matter when it gave the warning if they both avoided the collision? That time difference in the warning is most probably what causes the phantom braking and if you want the warnings earlier on the Mazda you can just change it to warn you earlier in the settings. As far as I know that’s not an option on Teslas

 

[SageBrush]

You’re comparing a $23k car

Someone else's choice -- I could not care less who the comparison car is, my point is that a similar rating does not make the two systems equivalent. The Tesla system is markedly superior, and if you cannot see why it would matter, I suggest you take your car out to a parking lot after a day of snow and drive around with summer tyres. See how well you can stop. Notice how long it takes and see for yourself if a delay of a second in starting braking makes an important difference

Or, just consider a little arithmetic: 37 mph is 60 kph. One second at that speed is 1/60 of a kilometer, so about 16.6 meters. All else being equal, do you think 16.6 meters in stopping distance is an important difference when emergency braking for a pedestrian ?

 

[sleepydoc]

A student at my daughters college was hit by a car and killed, at night, in the rain - no foul play/alcohol - just walking.

When the weather is bad on my boat I use RADAR, no other way.
When the evening arrives, otherwise clear, my tesla brakes for everything, and nothing.

The clock is ticking on the warranty expiring and gaining root access, someday the car might actually be mine.

No one can say if a system like Tesla’s, or any other system would have saved that student. Every system is worse at night, and every system is worse in the rain, whether it be radar, cameras or a pair of actual eyeballs. Walking at the side of the road at night in the rain is taking a significant risk; the goal of all of these systems is to provide another pair of virtual eyes and reduce injuries and fatalities. It would be foolish to expect them to catch everything and equally foolish to say they’re worthless because they’re not perfect.

 

[username123]

Someone else's choice -- I could not care less who the comparison car is, my point is that a similar rating does not make the two systems equivalent. The Tesla system is markedly superior, and if you cannot see why it would matter, I suggest you take your car out to a parking lot after a day of snow and drive around with summer tyres. See how well you can stop. Notice how long it takes and see for yourself if a delay of a second in starting braking makes an important difference

Or, just consider a little arithmetic: 37 mph is 60 kph. One second at that speed is 1/60 of a kilometer, so about 16.6 meters. All else being equal, do you think 16.6 meters in stopping distance is an important difference when emergency braking for a pedestrian ?

There you go fixed it for you. Unfortunately I am smart enough to put winter tires on all my cars so I won't be able to test that for you. In the test the Mazda would've been on the normal setting so now this should be on par with the Tesla's warning timing but without the phantom braking.

I just looked it up and there was a recall in late 2019 for the phantom braking but they were able to resolve it the year their system came out, something that Tesla still struggles with.

They both avoided the collision (Mazda did 3 years prior) so the 16m didn't matter but what does matter is that Tesla's system has false positives.

I suggest you do a little test to see why Mazda's system is superior, go on the highway (snow or not doesn't matter) and slam your brakes as soon as you see a shadow in front of you. Hopefully the car behind you was either paying attention or was driving a Mazda and avoided the collision. I know that I'm setting my Tesla's AEB system to late warning. It was on normal yesterday after only 400km of driving since I bought it and I already got a warning when the car in front wasn't even slowing down.