Featured IIHS automation safeguard ratings: only TSS 3.0 rated acceptable, Tesla and most others get an F

Your post is quite misleading.

This was not an IIHS test of how well the Level 2 ADAS works, but of how well the safe guards do their job. Like how well the driver monitoring operates. Though I don't see how requiring the seatbelts engaged in order for the systems to operate has much bearing. Maybe good idea, but so would calling for cars not to start without them fastened.

 

Independent from the seatbelts, personally I think the "Driver involvement" part is quite interesting. In the end it shows different philosophical approaches and it seems IIHS is now on the side of the Toyota philosophy.

Tesla was aiming very high (especially with promises) but it seems lately its falling more and more appart in certain safety/security areas that people happily ignored during the hypeyears. Whether its how you open the doors when sinking in a pond, how you can do things via UI without letting your eyes from the road or do simple thinks like honking or hazzarding fast in high stress situations.

That they cant provide proper auto parking feature without USS shows somehow they dont believe themselff in the safety of their custcotting features anymore... same with the cruise control, they just dont allow it for higher speeds.

 

My personal experience on Autopilot in a 2018 Model 3 loaner and my 2019 Model 3, version 2024.44.30.14, Monday, March 11, 2024,

• Driver monitoring - I get both steering and eyes on road alerts.
• Lane change - Navigate on FSD and AutoPilot initiates lane changes to follow route and move in or out of passing lane based on speed and local traffic. Some lane changes at intersections are late so I override for left turns.
• ACC resume - I always have to reenable. Sometimes I disengage to handle passing on two lane roads and difficult traffic.

I went to read up on their test protocol and found:

Safeguards for Partial Driving Automation
Test Protocol and Rating Guidelines
Version I
March 2024​

Scanning their test protocol, I suspect version "II" will be coming soon. Regardless of the obvious errors, it is a start.

Bob Wilson

​

 

No problem. The last non-Tesla driver assistance I used was on a 2017 Toyota Prius Prime and it was truly bad.

This is an area of rapid progress, usually about 6 weeks between Tesla updates. The next Tesla, AI based stack is expected to replace V11 with V12.

Bob Wilson

 

I think that in the future the only way to go is to put the steering wheel in the hands of something not human. But right now cars aren't necessarily safer than they were 30 years ago.

One type of death-by-car that has actually risen since the 90's are pedestrian deaths. Modernday vehicles are more likely to run over other people than cars did 30 years ago. Which drives me insane because I want to be able to walk to work or at least ride my bike, but I'm afraid to do so.

I'm glad that EV's have a lower center of gravity. I drove a Honda SUV yesterday (and didn't care to check what the name was, I'm guess Pilot, but am unsure) and like always, there's a warning on the sun visor of how the thing is easier to roll over.

 

surprise - surprise. not to worry - plenty of other articles out there to be skewed

.

 

I'd like to see the actual test results.

The lane change rating is just pass or fail based on whether the driver has to okay the lane change or the car can do so completely on its own. Their driver monitoring test times how long it takes for the alerts to start; can be as little as 5 seconds. Not having auto resume should be an auto pass, but maybe the 2021+ Model 3s have it.

Tesla has been generally slacking about these safe guards like driver monitoring.

I'm all for the safe guards, but being great in them doesn't equate to a great ADAS functioning.

I'd like to see use of universal terms for ADAS systems. I noticed the IIHS terms for lane keeping differed from Toyota's.

Some of that is cause of improving car safety. Stronger roof supports to prevent collapse in a roll over means wider A pillars that block the view of entire cars, not to mention people.

Is that warning requirement actually tied to a model's actual center of gravity or just car class? Does the Model Y or other EV 'SUV' also have it?

 

Also the individual attempts for getting into a "safer" vehicle (aka: SUV and pickup justification). Higher hoods also make it harder or impossible to see things in front of the vehicle. They also mean it's more likely to crash into the part of the body with vital organs, then push the person to the ground and then run over him/her, whereas a shorter vehicle and hood will likely do damage to the legs (which isn't as life threatening) and cause the person to fall onto the less leathal hood and windshield, causing less damage to vital organs and avoiding getting ran over.

 

Speculation: It's there for legal reasons, so it's probably tied to what the car/truck/mfv is legally speaking.

I'm hoping this gets better with more of these automated safety systems, because the dimensions of trucks are not friendly at all to busy mixed pedestrian/cyclist/automobile traffic.

To that end, it would be nice to know which systems are default / available in the most popular trucks: the Ford F-series, the GM fullsize trucks and the Ram pickup line. This document only lists GM Supercruise for the GMC Sierra truck.

I still think there's a lot to be done re: Vision Zero and stronger enforcement within dense low-speed areas is of great importance.

 

The brand new Honda I drove kept acting like I was going to get into a headon collision. It even tapped the brakes unexpectedly a couple times. Then today I saw that the NHTSA is investigating this problem saying it's even caused a few accidents.

My thoughts are:

1. Adding more and more accident avoidance systems eventually takes all the control out of the hands of the driver, which kind of takes the fun out of driving.
2. Maybe if we didn't put motorized vehicles where people walk and bike things would be better.
3. Is trying to perfect the personal vehicle the right thing to do? So many other forms of transportation are much safer overall and reduce or eliminate the problems with personal vehicles, such as safety, pollution, cost, etc.

 

I blame this on the improved rollover protection. To keep the roof from crushing the occupants, the "A" pillars on both sides of the windshields are thicker to be stronger. But you can't see through them.

During the initial 'Bell the Hybrid' nonsense, there was a bias in the data depending upon a left vs right turn for left hand drive and the mirror image for right hand drive. With the thicker "A" pillar closer to the driver, it obscures a wider arc and the biologically built-in blind spot makes it worse. The blind spot we all have gives the illusion the other eye can see everything but that is not true.

Solutions come to mind:

• Gorilla glass "A" pillars
• "Transparent" monitors
• Tesla wide area cameras on windshield, rear view mount

Bob Wilson

 

About the IIHS scores:


Bob Wilson

 

The problem started in the 1990's, long before smartphones even existed. I blame it on a list of contributors that have only made it worse:

• Thicker A-pillars.
• More things people stick to the windshield (e.g. phone holders, etc.).
• Preferrence for taller vehicles with taller hoods.
• Cellphone/smartphone usage.
• Dashboard dining, grooming, etc.
• Android Auto, Apple CarPlay and touchscreen interfaces in general.
• The stress from the pandemic.
• The stress from newer generations having fewer economic opportunites and older generations judging them.
• Poor street designs.
• Overpowered vehicles.
• Lack of sufficient driver's education.
• Lack of sufficient law enforcement.

I like the ideas. Also truss style A pillars that have glass in the gaps so you can see through them.