NTSB To Investigate Fatal Accident in San Antonio Involving Ford BlueCruise

[hturnerfamily]

the introduction to these newer 'automated driving' technologies is what is spawning all the chatter... it's less about the real issue, which is all the other 'elements' of the accident, which, if no 'technology' was part of the equation, would have not even made any news...

there are plenty of 'reporters' and 'writers' who want to create 'spam news'... it drives their business - no pun intended... or maybe pun WAS intended.... : ?

Sponsored

 

[Maxx]

I'm not saying anything is good or bad. I'm suggesting, in the aggregate, roads are safer with modern vehicles with advanced driver's aids. Do you agree with that statement or not?

Although I have not seen any data, my gut feeling says this is true.

What safety feature do you think is unsafe? BlueCruise?

Automated steering control as it stands today.

 

[ATLJoker285]

Slightly off topic since my post does not directly involve the tragic fatal accident, but it may address the potential to reduce future fatal accidents while using BlueCruise. A head's up ... most readers are not going to like this one ...

Why do BlueCruise and other driver-assistance systems allow the driver to set and maintain a speed which is above the posted speed limit?

Speed limits are supposedly set to a "safe" speed based upon typical driving conditions and accident history for a given stretch of road. It could easily be argued that the driver-assistance system designers are "assisting" unsafe actions by the driver. I am very surprised that the NTSB has not addressed this issue.

Since I was a child I've always wondered why vehicles are sold & allowed to go past any speed limit. If it's about regulation, cap capable speed at 70mph.

Then I grew up & realized, you can't regulate every single thing in this world. Once Pandora's box is opened.... Make duller knives, ban non-hunting guns, ban diving boards AND swimming pools, governors on all cars, etc. At some point, personal responsibility & accountability has to factor in.

 

[Maxx]

Since I was a child I've always wondered why vehicles are sold & allowed to go past any speed limit. If it's about regulation, cap capable speed at 70mph.

Then I grew up & realized, you can't regulate every single thing in this world. Once Pandora's box is opened.... Make duller knives, ban non-hunting guns, ban diving boards AND swimming pools, governors on all cars, etc. At some point, personal responsibility & accountability has to factor in.

His assumption that safe speed is the posted speed is incorrect. The safest speed is the average of traffic flow. the lower the speed differential, the higher reaction time to work with and a better chance of recovery in case of some collisions.

I used to be a traffic engineer a few decades ago. And we would get regular requests from politicians on how to put aside sound judgement to make them happy. Our director was nationally respected and wouldn’t put up with any crap. I am still surprised how long he was able to hold his job while protecting us so we could do our jobs. But I doubt that is the case everywhere. A lot of times, a traffic engineering feature is there because a soccer mom knew how to manipulate media or someone threatened to pull funding.

 

[Jim Lewis]

His assumption that safe speed is the posted speed is incorrect. The safest speed is the average of traffic flow.

I think you also have to consider that there are stationary objects along the road to be hit, such as the stockstill Honda CRV hit by the Mach-E in the news article or objects that have fallen off vehicles (my wife hit a paint bucket at 65 mph on a very busy Interstate where she couldn't swerve out of the way because of all the other traffic surrounding her; a neighbor's son hit a car jack lying on an Interstate, and his vehicle made it to an exit ramp and died on the ramp. He called his Mom, asking what he should do on the ramp with no safe way to walk away from his vehicle, and she called 911). You can't overlook the Newtonian K.E. = 1/2 m v^2. In other words, the kinetic energy involved in an accident goes up by the square of the velocity of the vehicles. An accident at 80 mph vs. 60 mph doesn't have just 1.6x more energy involved. It has 2.6x as much energy. So that's where the saying, "Speed kills..." comes from.

Also, the stopping distance after brakes are applied increases in proportion to the square of the speed, so stopping distance (and even just the lack of decrease in velocity up to the point of impact) will go up by the square of the speed. The time to react and apply the brakes also goes down because you're traveling faster and closing faster while you're "thinking" about reacting.

https://copilot.microsoft.com/sl/fDNP2WkGLds

The higher the speed, the more likely an accident will be fatal:
Speed (iihs.org). Traveling in a high-speed cohort doesn't insulate you from the basic laws of physics.
https://copilot.microsoft.com/sl/fwZSUe81SQm

I read some years ago that the V.A. Hospital prescriptions are filled by robots who put the pills in the bottles. The V.A. found them more reliable than human dispensers. Maybe the same will someday apply to robotic drivers vs. humans when vehicles have more powerful computers, better sensors, and better AI.

How are speed limits set for high-speed roads: https://copilot.microsoft.com/sl/tNEDKUCavQ

 

[Jim Lewis]

The Associated Press seems to be the original source of the story "NTSB To Investigate Mach-E Crash into Honda CRV": US to investigate Texas fatal crash that may have involved Ford partially automated driving system | AP News.

That article seems to be written with less hype and simply states NTSB's reason for being involved:

“NTSB is investigating this fatal crash due to its continued interest in advanced driver assistance systems and how vehicle operators interact with these technologies,” the agency statement said.

 

[Maxx]

I think you also have to consider that there are stationary objects along the road to be hit, such as the stockstill Honda CRV hit by the Mach-E in the news article or objects that have fallen off vehicles (my wife hit a paint bucket at 65 mph on a very busy Interstate where she couldn't swerve out of the way because of all the other traffic surrounding her; a neighbor's son hit a car jack lying on an Interstate, and his vehicle made it to an exit ramp and died on the ramp. He called his Mom, asking what he should do on the ramp with no safe way to walk away from his vehicle, and she called 911). You can't overlook the Newtonian K.E. = 1/2 m v^2. In other words, the kinetic energy involved in an accident goes up by the square of the velocity of the vehicles. An accident at 80 mph vs. 60 mph doesn't have just 1.6x more energy involved. It has 2.6x as much energy. So that's where the saying, "Speed kills..." comes from.

Also, the stopping distance after brakes are applied increases in proportion to the square of the speed, so stopping distance (and even just the lack of decrease in velocity up to the point of impact) will go up by the square of the speed. The time to react and apply the brakes also goes down because you're traveling faster and closing faster while you're "thinking" about reacting.

https://copilot.microsoft.com/sl/fDNP2WkGLds

The higher the speed, the more likely an accident will be fatal:
Speed (iihs.org). Traveling in a high-speed cohort doesn't insulate you from the basic laws of physics.
https://copilot.microsoft.com/sl/fwZSUe81SQm

I read some years ago that the V.A. Hospital prescriptions are filled by robots who put the pills in the bottles. The V.A. found them more reliable than human dispensers. Maybe the same will someday apply to robotic drivers vs. humans when vehicles have more powerful computers, better sensors, and better AI.

How are speed limits set for high-speed roads: https://copilot.microsoft.com/sl/tNEDKUCavQ

I don’t see any conflict between what you said and what I said. I agree that if you can bring the average speed of the traffic 10 mph lower than speed limit, instead of 10 mph over speed limit, the severity of accidents will be reduced. But if your point is you driving slower is always safer, you must be doing 20 on interstate while everyone else is doing 80 and consider that a safe behavior.

No one here needs to know about physics of momentum to know an accident at higher speed is likely to do more damage. It is common sense. What you are missing is that when you are doing 20 while everyone else is doing 80, You significantly increase the chance of getting hit with 60 miles an hour speed differential. If you were doing 80 instead, even if speed limit was 70, you reduce the chance of collision with other cars around you.

We had a cow jump off the bridge falling on a car killing the entire family. would you pay more attention to flying suicidal cows more than cars around you? What makes a behavior safer or more dangerous is not just how bad it would be when the accident happens, it is that multiplied by the possibility of the accident happening.

As I told you I was a Traffic Engineer, we didn’t pull speed limits out of our booty. We always did shoot for zero fatality on Maryland highways. It was an important performance measure. We had an entire office dedicated to accident studies, working with state police to translate all new information to better designs. We had visitors from Brazil and Russia to learn from us. But highway performance; how many cars we moved and how much time an average individual wasted on our roads were another performance measure.

We did all kinds of traffic calming methods because we knew the fatality rates for bicycle accidents in European cities are much lower than it is for car accidents on American roads, but you are not going to swap your Lightning for a bicycle, are you? So we work with reality to make it as safe as possible.

All I am saying is you want your speed as close as possible to the cars around you. If you are on a 6 lane highway and traffic is slower on slow lanes, all your points stand compare to fast lane. However if you do decide to do 20 at 70 because you think slower is always safer, please let us know where and when you are driving.

Edit: I have no doubt robot drivers will be safer than humans at some point especially when there are more of them on the road than we are. But today, they not only have navigate stationary physical world, they will have to deal with unpredictable human mind moving other cars. Now that machines are collecting data on every little move we make, we will be a lot more predictable soon. It will happen but as Musk has realized, it may be a little more challenging than he initially thought.

 

[Jim Lewis]

@Maxx, to put my remarks in context, I lived through the 1973 Arab Oil Embargo, and very expensive gas (for the day) and gas lines in 1974, etc., resulting in the reduction of the national speed limit on Interstate highways to 55 mph. One of my favorite songs coming out of that experience was Sammy Haggar's I Can't Drive 55:


But I'd say to make your point, you really go to an extreme, hypothesizing someone driving 20 mph on a designated high-speed, limited access road. On such roads, you can actually get ticketed for driving too slowly. There's usually a minimum allowed speed on Interstates, set separately by each state and territory, not to mention the maximum allowed speed limit.

To theorize that the safest speed is the speed everyone else is driving because otherwise, you'll cause an accident is just plain silly. There are many other factors that go into determining safe driving speeds regardless of what the herd behavior is. If curves are built for 65 mph, driving them at 75 mph on slick roads because everyone else is doing it makes it magically safe(r)?!

How are speed limits set for high-speed roads: https://copilot.microsoft.com/sl/tNEDKUCavQ

Basically, it all boils down to safety recommendations, which effectively are what speed limits are all about. However, speed limits and actual driving speeds are not always determined solely based on safety; social, political, and economic considerations also factor in. It really almost becomes mystical when somehow, every day out on the road, a group of drivers hurtling down the Interstate can collectively arrive at the optimum speed to drive that day on that road with no real reference to highway engineering, accident history for that stretch of road, what's ahead around the curve in the next few miles, construction, weather, etc. Just do what everyone else is doing now, and you'll be as safe as you can be. Perhaps Darwinian evolution will ultimately provide us with the true answer in a few centuries or millennia as unsafe drivers keep killing off their family line, taking their wives and kids with them in high-speed accidents. We'll gradually evolve into a safe-driving species and have the instinctive knowledge of whatever safe driving is truly in our genes.

 

[Willebrew]

This isn't a radar issue, radar is perfectly capable of detecting stationary objects.

The issue is with how vehicle manufacturers process the data from the various sensors, ie short and long range radar as well as cameras, to determine what is relevant information and what is noise. On a side note, it would be really cool to see the raw radar data from the Lightning.

I thought Tesla was still using their Tesla Vision for everything and no longer using radar or lidar, or have they switched back?

You are correct, the radars in the vehicle can detect stopped objects, and the system has to pick and choose what objects to react to, and what to ignore (such as speed signs, traffic lights, overpasses, etc.), and obstacle recognition is a know challenge with radar data (radar resolution is a limitation). This is why BlueCruise fuses camera and radar data, like what Tesla used to do, to combat this challenge. Unfortunately, in this case, the CRV didn’t have its lights on, it was stopped in the middle of the road, and considering that BlueCruise was traveling at highway speeds, it doesn’t come as a surprise that this happened (the camera most likely had trouble seeing the stopped car since it was dark, so BlueCruise couldn’t use it to validate the radar data). What Ford can do is take all of the data logged from this accident (BlueCruise, like Autopilot, logs a ton of useful data), create simulations of similar situations to train BlueCruise on, validate the improvements and then push the improved version of BlueCruise to the fleet. After all, the developers at Latitude AI are very proud of the Continuous Learning Loop (announced with BlueCruise 1.3), and this is a great way to use it.

 

[RickLightning]

Here is my issue with it:

How many of the folks that use the system are in fact FULLY attentive during the entire trip while they use the system? And of those (if any) how many see an advantage in using the system? My suspicion is the only advantage is when you are NOT fully attentive.

That of course requires honest answers not just from those that want to protect their precious toy.


p.s. Adaptive cruise and lane centering I could relax about a little if they worked reliably. My lane centering has not been reliable. I only turn it on when I am really in bad shape (Exhausted).

If you are in "bad shape", why are you driving? You are putting yourself and others at risk. Isn't that irresponsible? Wouldn't it be good if the vehicle's systems detected this, and warned you, and then stopped the vehicle if necessary?

Wait, wut? That is exactly what they do? Wow...

whether you are right about this or I am is a matter of total impact and total difference with and without BC. If Tic Toc guy was spending 2 minutes of a 30 minutes drive looking at his screen before he got BC and now he is spending 15 minutes, the total impact still can be negative. What I am suggesting is that regardless of how distracted you were before, you will likely be more distracted with another driver at the wheel.

The problem with these technologies is that they are great at 99% of the situations 99% of us usually are. And when that 1% happens to someone else, it is just a news about some unlucky schmuck. Our personal experience has been 100% positive so we get more confident and we use the system more until statistically, that risk is no longer insignificant to us as individuals.

p.s. You can have a lot of safety features without the car driving itself. It is not an all or nothing deal.

Clearly you lack understanding of the systems in your truck. Find a piece of road and go driving with someone. When the road is empty and straight, trying using BC, handsfree or not, while looking at the screen...

 

[RickLightning]

brakes...

braking...

pedals...

People have a responsibility to understand the capabilities of the technology that are using. Your vehicle has many capabilities, yet people don't read manuals, nor understand limitations. RTFM.

The manual clearly tells you what these capabilities can and cannot do. Imagine driving down the road where a vehicle is parked on the shoulder, on a curve, lights out, Your vehicle looks straight ahead and there is a dark mass there. Should it slap the brakes? No, that is the driver's responsibility.

IMO, this driver is going to get charged with a crime for killing someone. Ford will be free and clear.

 

[Heliian]

IMO, this driver is going to get charged with a crime for killing someone

Unlikely, the car was stopped with no lights at night on a highway. The Ford driver won't be charged with a crime.

 

[Maxx]

To theorize that the safest speed is the speed everyone else is driving because otherwise, you'll cause an accident is just plain silly. There are many other factors that go into determining safe driving speeds regardless of what the herd behavior is. If curves are built for 65 mph, driving them at 75 mph on slick roads because everyone else is doing it makes it magically safe(r)?!

It does. But it only sounds magical because you may not have worked with the herd before. Herd mentality is way underrated. The posted speed limits, the road conditions, the capability of each individual and each car, ……. all of those factors has an impact on the herd behavior. There is a reason herd mentality has ensured survival of many species for thousands of years. Don’t knock it down until you have tried it .

Human Factor Engineering goes hand in hand with designing for safety. If you see folks consistently drive 10 mph over speed limit, you factor that in, in your design. It is always a back and forth between how the herd respond to a design and adjust accordingly.

Don’t forget, you are a part of the herd so on a one lane road with a posted speed limit of 40 mph, if everyone else wants to go 50 and you decide it is not safe and go 40, the herd will end up doing 40. Of course that may cause some folks lose their mind and engage in dangerous passing that increase the chance of fatality all because you tried to stay “safe” but that is an entirely different can of worms. What I am saying is any system that involves a lot of humans making decisions, is a complex system and you can’t reduce it to a speed limit sign.

I am not going to beat this to death. I do agree with you that we should just let natural selection do it’s thing but here are a few pieces of information from Federal Highway Administration indicating, they don’t think it is that silly:

 

[Jim Lewis]

I am not going to beat this to death. I do agree with you that we should just let natural selection do it’s thing but here are a few pieces of information from Federal Highway Administration indicating, they don’t think it is that silly:

Your particular citation doesn't mean a whole bunch without providing information, such as how much the average speed differed from the posted speed. There are also no error bars on the data on the graphs. If the average speed deviated little from the posted speech, the "data" you post would essentially show that it's safest to drive the posted speed. The first graph is data from 55 years ago, three generations previous to the current. Many of the drivers who provided that data are probably deceased or at least not driving very much today.

 

[Jim Lewis]

IMO, this driver is going to get charged with a crime for killing someone. Ford will be free and clear.

Since the articles announcing NTSB involvement appeared a couple weeks after the accident (2/24/24), and the driver's name is not mentioned in either the MySanAntonio.com or the AP article, that's usually an indication, at least in San Antonio, a person has not yet been charged with a crime. Quite possible the driver could be charged, though, as more details become known. The Honda CRV driver could have been dead drunk, dead at the wheel, or just the victim of a sudden engine failure. In the 1980s, a Nissan Stanza that I loved blew a head gasket traveling at city speeds and became a dead-in-the-water vehicle within a very few seconds. I then discovered head gaskets weren't considered part of the drive train in the extended warranty I had bought from the dealer, rather just a normal wear-and-tear item!

Sponsored