A smorgasbord of safety
You try to come up with clever titles each week. Sometime they work and sometime they are a, “smorgasbord of safety”.
We discuss some extra CES goodies, thermal cameras, AEB, driving simulators for teens and rats. Fred discusses a recent Microsoft paper(pr?) on using AI to find better battery chemistry.
This weeks links:
Subscribe using your favorite podcast service:
note: this is a machine generated transcript and may not be completely accurate. This is provided for convience and should not be used for attribution.
Anthony: You’re listening to There Auto Be a Law. The Center for Auto Safety podcast with Executive Director Michael Brooks, Chief Engineer Fred Perkins, and hosted by me, Anthony Simenon. For over 50 years, the Center for Auto Safety has worked to make cars safer.
Fred: Well, good morning again, gentlemen.
Anthony: Yeah, greetings listeners. Thanks for joining us again. We’re gonna do another little follow up from last week on CS, the Consumer Electronics Show. Where is BMW had this cool little demo, well it sounds kinda cool, where it’s remote valet. Where this is, this is something they talk about, like, hey, you, you, you go to your fancy restaurant or something like that, and you don’t wanna hand your keys over to some good, somebody who’s wearing a red Vest.
Instead you park your car there and then remotely the car will park itself and you’re like, that sounds amazing. And at first I’m thinking, wait, have they managed to figure this out? Now, what they’ve done is they’ve got somebody sitting in the Philippines remotely driving your car to a parking lot.
There’s, I don’t, who is asking for this? Like, I would rather pay some teenager in a red vest to go ahead and park my car. Cause I can have a conversation with him.
Michael: Well, don’t forget we’re talking about BMW drivers here. They’re a special little class of no flakes. And, you know, I don’t, I don’t know. I mean, obviously it would save time.
It was, it was kind of, it’s, it’s scary to me for, I think, one reason. I’ll let Fred talk about that, but if you’ve got a guy controlling a vehicle from the Philippines, you’re going to have latency and you’re, you’re, it’s, I just don’t see how they’re going to be able to make that safe with remote driving.
Anthony: mind, I just made up the Philippines as a location. It’s going to be a remote location, but I don’t imagine it’s going to be somebody. You know, within spitting distance of the restaurant you’re parking at.
Fred: Imagination is good. That’s good. We like that.
Anthony: So, Fred, besides, I just imagine like most parking garages I’ve been in, there are these large concrete and steel enclosures.
Radio waves don’t really like concrete and steel so much. So, do you see any problems with this setup?
Fred: Well, it depends on who you’re talking about. If you’re talking about the emergency response people, yes. If you’re talking about the, Insurance industry, yes, if you’re talking about the drivers who are going to save, oh, I don’t know, 30 seconds or so by trusting their car to somebody they’ve never met with spotty connectivity, I suppose that’s going to be just fine.
I, I don’t. Are people really that busy that this is an important thing to do?
Michael: Yeah, I, I can’t figure it out. Same thing with the Tesla. The Tesla had something similar where it parks your vehicle in a parking lot, which is just, that is, you know, the height of laziness or, you know, creature, creature feature, you know, I don’t understand why anybody would need to do that in a small parking lot where, you know, it’s easy to park, generally easy to park somewhere.
And find your way into the business. And that’s 30 seconds here. The BMW system seemed more geared towards city environments where you’re pulling up to a restaurant and you don’t want to circle the block 20 times looking for a parking spot before you, before you go in. Or you don’t trust the, the, the valet at the restaurant or the hotel or wherever you might be.
Be, so it’s probably a little more of a complicated situation. I was, I, when I was first reading the article, I was, I, you know, I was a little concerned that what they were talking about was a self-driving type of parking function. But it seems like it’s purely going to be a, you know, a remote, you’re gonna have your own little remote driver that you pay a subscription to, to, to, to.
Carry out the parking of your vehicle which as we, every time we discuss something like that, where there’s a remote driver, the latency issue comes up and seems to be somewhat of an insurmountable problem.
Fred: Well, if latency is the only problem, that’s a good day. You know, if I were, if it were my car, I would, you know, want to check out the parking lot ahead of time to make sure there was a space for the car, because I wouldn’t want it aimlessly roaming around under the control of somebody whom I’ve never met, who I have never qualified.
So how is this saving time? If you’re going to check that there’s a parking spot ahead of time, how are you going to do that? You’re going to drop your car, walk around the parking lot, and then come back to your car and push the button for the automatic driver? This I don’t, I, I, It’s a wonderful thing, I
Anthony: suppose, and It strikes me as being very anti social, but I, you know, you’re a BMW driver, what can I say?
Well, I’m somewhat, off a slight tangent from this, I see this as, as kind of what I see with a lot of use cases for self driving cars is, they say, hey, get into your self driving car, you can go fall asleep, and then you’ll wake up at your destination. And in my mind, I’m like, that’s the bus, like, that’s a long distance bus ride, like, I can go in there, oh, I can watch movies, I can eat snacks, I can use a bathroom, it’s great, the seats recline, somebody else is driving it but hey, I’m old school, I guess.
Fred: I don’t know. BMW must think it’s important. They’ve spent time and money developing this, but it’s a mystery to
Anthony: me. Yeah, so BMW drivers out there, yes, all kidding aside, if this is something you’re like, yes, I want to pay someone remotely. to drive my car, like, and they’re set up like it’s a video game system.
So the remote drivers basically have high end video gaming rigs to drive your car. Granted, they can’t see 360 degrees like a human does in a car. They can see whatever display they’re seeing. But if you think that, hey, this is something I want to pay. I want to pay a monthly fee for this instead of, you know, randomly, you know, tipping some guy five bucks let us know, please write in contact at autosafety.
org. I’m dying to know. I won’t make fun of you too much, maybe just a little
Fred: bit. Well, I’m a big fan of irony myself. And so for a company that advertises itself as the ultimate driving machine, what they’re really saying is once you get to the restaurant, you’re going to be sick of driving this damn thing.
So turn it over to somebody else. Uh, I, I think the engineers ought to talk to the marketing people and try to get a coordinated approach. .
Anthony: I lo I love it. I love it. Also out of CES another German company called Mercedes, they have something called, it’s the g the their electric G GWA is capable of doing what they call a G turn, which they say must not be used on public roads.
And it’s kind of like this. Four wheel turn, so you can do this 720 degree super tight turn, which is kind of neat and interesting. I mean, it’s, but, one, hey, here’s this feature you’re paying a lot of money for. You cannot use it on public roads. I don’t know who’s clamoring for this again, but you can see images and a video of this.
We’ll link to the article in Jalopnik, where, is it just another, is this just a big tail fin on a car from the 50s?
Michael: Yes, you know, I, I know what it, the only folks that are really buying these Mercedes G wagons are, I would say they’re primarily the look at me set. So, yes, those are the people who would be using a G turn, which is also called a tank turn.
Basically, it’s just. Your vehicle is facing North. You hit a button and the vehicle rotates itself to face South. Now you could see situations in which that could come in handy. It, you know, it prevent you from having to do a Y turn. If you went down a one way street on accident, right? There could be uses for it.
But, you know, the way that Mercedes has presented it, you know, with Three or four of these vehicles with disco lights and dancing some way kind of, I think Kate, they don’t really mention the actual use cases for it. They’re, they’re, they’re advertising it for it’s, you know, flashiness and how cool it is that a car can do that.
But the real problem here is one that, you know, we’ve seen before and probably mentioned with the new Hummer EVs is that they’re, they’re putting a feature out and it’s, it’s potentially unsafe if used improperly, not nearly as unsafe as the Hummer’s WTF mode, which is a zero to 60 tank in three seconds going down a public road, which is concerning.
Here, you’ve got a vehicle that’s basically just. Turning on its axis in the middle of a road but you, you can’t tell people not to do something. They’re going to do it. You can’t put out a car and say, you know, do not use this on public roads. They’re going to do it. They’re going to use it there. And.
If these G wagons are connected to the internet and they know where they are, you can geofence these things. You can ensure as a company that it’s not used on a public road and it’s not expensive to do so. My question continues to be on these little weird performance features. This one’s Not as concerning as some of the others, some of the super speed launch modes that we see across the industry, particularly in a lot of the new EVs, where they’re going to be able to achieve just obscene acceleration, you know, in the middle of a city, wherever you want to do it, that button is there.
When those types of features. Should be, you know, actively prevented from being used when you’re in an area where they weren’t designed to be used. And that’s something that we think is going to become more and more important. We’ve seen a lot of issues with the Tesla is not being geo fence properly turning autopilot or full self driving on areas where it’s not designed for tested for to be safe.
And there continue to be a lot of these. Odd little features that are trickling out that manufacturers saying, well, well, we’re going to put this in the car and you can use it at any time, but don’t, which I think that we all know that that humans don’t really work that way. At least a lot of us aren’t able to take those warnings in and and do the right thing.
Anthony: Yeah, the only modern feature people don’t use in their cars is the AM radio. Because how do you even get to that? I have no idea. Alright, let’s let’s jump to some actual safety directly related to safety issues. Here’s an article we’re linking to from Electronics 360, my favorite electronics magazine.
It talks about why automobiles need new nighttime safety systems. Quoting from the article, Simply put, data from the Federal Fatality Analysis Reporting System shows that as cars have gotten bigger and drivers have become more distracted over the last 10 years, pedestrian fatalities have greatly climbed, with over 75 percent of these deaths occurring at night.
The New York Times had an of an article a couple months ago, maybe it was last month, talking about all of this increase in pedestrian death and, I don’t know, on safety? On safety, sure. Lack of safety? It’s more dangerous to be a pedestrian today than it was yesterday.
Fred: This particular article was interesting.
It was written by a company that’s selling thermal imaging systems. So there’s a bit of a bias there, but it does point out that there is additional sensory information that could be brought to bear for vehicle safety. And I think that the combination of the however limited thermal imaging systems that are available now with the other sensors that are on the vehicle could be used very well to reduce The deaths at night, pedestrian deaths at night sensor fusion is probably a quicker approach than perfecting the system, as they talked about it in this particular article, but yeah, there’s a lot that can be done.
Michael: And we’ve, we’ve looked at thermal imaging somewhat, and we’ve talked to a few folks in the industry about it. And, you know, it seems very promising. I mean, you’re, you’re not just able to detect. Humans, you can detect, you know, you could build a car that never runs over a dog again. You could do a lot of things with this technology to detect deer, which cause all sorts of problems on our roads every year and, and other things that could, you know, really improve safety for vehicles.
I think. One of the areas, and this article is, you know, pretty transparent about the challenges that are, that are being raised here, not just with the cameras, but also with something we believe is that the biggest hurdle here, which is, you know, how do you classify you know, having the sensor and detecting certain objects and detecting things is one thing, but how do you On the back end, create software to classify those objects appropriately and make sure that, you know, you’re not slamming on the brakes at 75 miles an hour for a rat that crosses the road.
We’ll get to the rats later and why drivers might not but there’s, there’s a lot of things that, that, that. Really need to go into this to ensure that we’re not seeing some of the the problems we see with other systems Like we don’t see that, you know, this is a system that could if it doesn’t detect something right cause a phantom breaking incident Could create more safety problems than then it would solve unless these they’re you know There’s this equipments calibrated properly and it has some really good software that can Distinguish objects from one another so that you’re, you’re, you’re, you’re getting things right.
Anthony: Is it an expensive system to add?
Michael: You know, we, we’ve heard figures about two, around 200, I think they were for, for these systems. And that was maybe two years ago. So who knows if they’d come down. So that’s less than the
Anthony: cost of formats.
Michael: I think it’s pretty well. You must have some bejeweled formats, but it seems like it’s within reach.
I mean, 200 is a lot, really, when it comes down to it to for auto companies to invest in a technology. They really like to be under that, I believe, for some of these, but it’s one that could really make a difference. And it’s one that I think this has been one of those areas where, you know, with automatic emergency braking, we’ve seen a lot of problems.
And challenges at night for these systems to detect other vehicles and particularly pedestrians and low light. And, you know, if a system like this is, you know, cost, if the cost is there, if you can get it in there at a point where it can be put into all the vehicles at low cost, then it makes a lot of sense to solve that problem.
The only issue there is that. The NHTSA is probably going to be hesitant to specify that companies use a specific technology there. They try to remain somewhat tech neutral in their rules, so I don’t know if they will require specifically this infrared or thermal camera technology, but they could, you know, set set up a rule that requires vehicles to perform at levels that maybe only a thermal camera could produce, but who knows how that’s going to play out.
It’s, that’s going to be in part two of the AEB rulemaking by NHTSA. I don’t think they’re going to address that issue in the rule that we hope is going to come out in a couple of months this year.
Anthony: Let’s continue talking about automatic emergency braking. The Insurance Institute for Highway Safety they talk about it’d be great if these systems could detect large trucks and motorcycles better.
Front crash protec pre blahhh Front crash prevention systems aren’t as good at preventing crashes with large trucks and motorcycles as they are crashes with cars. Two new studies from IIHS show. Which, I mean, the motorcycle part I can understand, cause it’s a relatively narrow object in some of these systems aren’t, their field of view is not that great, but a truck is the one I don’t understand.
These things are massive. Like, how are they missing a truck,
Michael: Tesla? Well, people say the same, yeah, exactly, Tesla. I mean, you say the same thing, but then, you know, Teslas are going under semi, I don’t know if it’s the height of the truck. Maybe there’s, you know. Maybe these systems have been programmed, you know, traditionally to detect cars only because, you know, 10 years ago when they started building this systems, it was really thought of as a way to prevent crashes between vehicles rear end crashes.
It’s it’s over time. It’s developed to where we want the technology to address the pedestrian issue. Which is a little more difficult and we want a broader range of vehicles covered. So that would be motorcycles and heavy trucks and anything. Really? We want to detect any hazard in front of the vehicle trains.
It doesn’t matter what and slow and slower. Stop the vehicle when necessary. But it’s important here because if you think about the, the, there’s a vehicle that needs to send something in front of it. If it, if it’s a, if it’s, if it’s a motorcycle, that’s missed, you’re going to have a far higher chance. If that motorcycles is not detected of seeing a death or a serious injury, because you’ve got a car crashing into the back of a motorcycle, which has relatively little protection.
And also, if you’re not. Yeah. Picking up the heavy trucks, then the car is going to be presumably sliding underneath the rear of that truck, which is is behind a lot of the fatal rear end crashes we see. So, these may be while while they may happen less often than the car to car scenario. The risks and the potential injury or the outcomes involved are, are much worse in these circumstances.
So it’s very important that as NHTSA finishes up its AEB rule, that it addresses these situations because they are two situations where we see, you know, a lot of carnage on the roads.
Anthony: One of these AEB rules supposed to be finalized. Well,
Michael: they put out the rule, the, the notice of proposed rulemaking last year, and that’s been commented on, and, you know, at some point this year, it, it, it’s going to be finalized, hopefully, you know, sooner rather than later, because the later you get in, in an election year, the better chance that an incoming administration is going to kick a rule out, essentially an incoming, you know, I would just, you If you have an incoming, you know, Democrat administration next, that rule is most likely going to stay as is if, if it’s, if it’s put out, you know, right before the new administration comes in, say it’s comes out in October, November, there’s a chance that if a Republican administration comes in, they’re going to say, Whoa, we don’t want, we don’t want any of this, throw it out.
We’re going to start over. And you know, it happens both ways. NHTSA did that. The, the, the Nits under the Biden administration did that to at least one or two rules that the former administration was trying to put into effect when they came in. So it’s something that, that if the sooner the rule gets out this year, the better chance it has of staying
Anthony: in effect.
Yay. Let’s I’ll try and make automatic emergency braking better because from personal experience, it’s okay.
Michael: It works. I mean, look, the good thing about this study and one thing it showed is that, yes, you know, there’s a 53 percent reduction here and rear and crash weights with passenger vehicles.
There’s a 38 with heavy trucks, 38 percent reduction there and 41 percent with motorcycles. That’s just not high enough. You know, we want to see 99 percent 100 percent there if possible. I doubt it. But. you know, as close to perfect as we can get. And that’s, that’s, it’s working right now. You know, IHS looked at 160, 000 crashes in part of this study and, you know, that’s, that’s good news.
That’s really good news.
Fred: I want to remind our listeners that there are no rules in place right now for automatic emergency braking. So the benefits that Michael just talked about have. been or significant, but they have not been as a result of any stringent rulemaking or any standardized tests by the government or any way of qualifying which are good and which are bad.
So there’s a lot of room for improvement and where we are now is not where we should
Michael: be. Yeah. And I think we’ll see that improvement once we have some standards in place when, when you don’t have the discrepancy between one manufacturer and another, they have to meet certain similar standards. Plus there’s some things in the rule that are, that are really going to help out with some of the problems we’ve seen with automatic emergency braking, like with phantom braking and other issues.
Anthony: So I’m going to ask you guys a question, Fred, what’s the scariest thing you’ve ever done in a car? Yeah,
Fred: I can’t remember her name uh, well, I don’t know, I guess to be G rated I would say the time I spun out in a Volkswagen Bug going down a hill with a car full of all high school students who I had encouraged to skip that day.
Anthony: pretty bad. Okay, Michael, what’s the scariest thing you’ve done in a car?
Michael: You know, I’m having a hard time finding, figuring scary
Anthony: out, I mean. There’s a correct answer to this in your book. Fred’s wrong. Fred’s close, but not quite right. Michael, come
Michael: on. You know, I’ve, I’ve driven in some hairy situations before where I probably would have been better off not being on the road in storms and otherwise, but I’ve been fortunate not to be in any significant crashes or, you know, I’ve had a couple of situations riding with other people where they’re driving.
Put me in fear of my life. But no, I don’t know what the obvious answer is that you’re going.
Anthony: Kind of close there. The scariest thing both of you have done and I’ve done as well is teach a teenager how to drive. Oh, come on. Think about it. And this is pretty neat. There’s an article we’re linking to from the Wall Street Journal.
Whereas, it’s researchers at Children’s Hospital Philadelphia built a virtual driving assessment essentially a realistic car simulator, and found that it can accurately predict crash risk in newly licensed drivers. Why did the hospital do this? Because driving is one of the most important healthcare issues for teenagers.
I love this idea. The article talks about some of these kids, they got behind there and they killed some pedestrians. I’m much happier to have them kill them in a simulation than in reality. And if this could be a good little training step, I, I think, you know, I’m all for it. It would make me breathe slightly easier when putting my teenager behind the wheel.
Yeah, I think it’s a great
Fred: idea. My, my granddaughter is learning to drive right now in the Philadelphia area. And they’ve looked for this and have been unable to find it. So it sounds like a good idea, but it’s not yet deployed in any meaningful
Michael: way. Yeah. And I Children’s Hospital of Philadelphia, or we always call it CHOP, they’ve been behind a lot of youth and child safety initiatives over the years.
So this is the most recent one. And this is, this one’s really interesting. They took, it looks like they’ve had this simulator going since about 2017, so about six years ago. And they went back. What they did basically was went back and looked at the records of all of the kids who had and teens who had gone through this program and then track their accident history on the road as they drove through their first few years.
Match that up and concluded that. Yeah, you know, this system work. We were able to identify, you know, the good kids or the kids who didn’t have that many problems had a 10 percent lower crash rate than average. And they thought to in the, in the simulator didn’t, didn’t, didn’t do very well. And with the high risk candidates, they had an increased chance of being in an accident by, by about 10%.
So it’s something that looks like it could work and maybe give, give some parents a little peace of mind as they put their children on the road. I doubt it’s going to help them lower their insurance a whole lot.
Anthony: Wow. Okay. Yeah. The insurance I understand, but I love this. Cause I think about it from like, you know, no one puts, you know, a pilot just right behind the, The controls of an Airbus A330 and says, go for it.
No, they’re logging a ton of simulator time ahead of time before, before they let them out on the, I don’t know, the airways, the airways, the skyways the right ways, the road. Well, they don’t really put a lot of. Airbuses out on the roadways Um kind of defeats the purpose, but i’m all for this. I think this is a great idea listeners if you’re for this too, let us know let’s go to you know something a little grosser now.
You guys want to talk gross. Let’s talk about rats again We’ve talked about this before, where rats are eating people’s cars all over the place. Is it the soy plastic wiring? Because, you know, rats, they’re, they’re generally very, you know, health based, health conscious. They’re like, oh yeah, yeah, let’s I gotta gnaw anyway.
And I, you know, that petroleum based wiring insulation, not really my thing. I’ve become vegan as a rat. So, oh yeah, it’s soy wiring. Another article from the Wall Street Journal talks about this poor guy and In New York, who’s car keeps getting eaten by rats people go crazy, they’re putting hot sauce all over their cars and they’re wiring every day, they’re trying to move the location of their car, bars of Irish Spring in the cabin, I don’t understand it they even suggest peeing next to your car to prevent the rats from doing this, but I’m like, eh, I mean, at what point do you just like, I have no dignity anymore, and just let the rats win?
I What is, do we know what’s actually happening here?
Michael: No, well, I mean, we, I think we kind of do, I think we kind of do, we kind of got at it when we discussed it before is that, you know, those little suckers will chew on anything. Right. And they don’t care. And, you know, they, they get in your engine for warmth sometimes and they look up and there’s who are and they say, Hey, I need to chew because I think it was that, you know, that.
I think it was the rodentia was the, the family or the order of animals. They include squirrels and rats and all sorts of things. And in order to keep their teeth sharp, they’re constantly gnawing on things. Their teeth, I think their teeth grow throughout their life. And so they, they kind of constantly have to gnaw on things to keep their teeth in shape.
And so wherever they are, whatever, whatever wires next to them, whether it’s soy or rice or whatever you’re making out of them, they’re probably going to gnaw on it. So. I mean, the guy in the article though, I mean, he, he really went. Goes to extremes, you know, he, he doesn’t park in consecutive spots, like, he makes sure that his car is constantly a moving target to avoid, to avoid these little critters, so.
But they hide in
Anthony: the article is, is that his car is actually the Oscar Mayer Wienermobile. No, it’s not.
Michael: I mean he has really had to make some lifestyle changes to it to avoid these suckers. So it’s, it’s, you know, it looks like it’s probably if you’re in a, you know, if you have a house and you have a garage and, and you, you know, you have decent pest control, you could probably avoid this issue.
But if you’re driving around the city, And there are trash cans and restaurants and, you know, a higher population of rats that you might encounter in other areas, then it can be a real problem. And it’s a problem, not just for drivers, it’s a problem for insurance companies because they’re covering some of this damage.
And it’s. You know, it’s, it’s a problem all around, and I don’t know, and I don’t know that anyone has figured out a, a really good way of solving this one, other than maybe putting wires in places where they aren’t accessible to rodents, which ultimately is going to And it’s not, you know, an ideal solution for manufacturers to deploy.
And there, you know, it probably have to pass the cost on to consumers, but you know, these, these can cause safety issues, you know, especially when we have more electronics going into cars, more wires makes you wonder if they should start protecting those wires with some type of, you know, covering, whether it’s metal or.
Kevlar or something else that’s impervious to rodent
Fred: teeth. I think that the companies are now providing optional services for, you know, like we talked about earlier, the remote parking services, I think it’d be a natural extension to just go with some people walking carriers around the street. Terriers are really good at this, and I think it’s minor subscription, modest subscription fee for, you know, some students.
This guy’s a professor in New York and be a good work study program to, you know, have students walk terriers around the neighborhoods. I think there’s a lot of potential there. Solve a lot of problems at once. Employment, you know, people say there’s no jobs for kids. Well, there you go. And terriers are cute little dogs, and I, I think there’s a natural.
Natural fix for this.
Anthony: So, terrier dog walking slash anti rat patrol. Absolutely, yeah. You got my vote. I’m on board. Alright, let’s shift gears a little bit. You guys remember Dieselgate? Of course you do. This is when Volkswagens like, Hey, let’s go put our cars into environmental testing or whatever for emissions testing and when we know it’s there, let’s turn on some software to make it look like everything’s good.
But when you’re actually out on the road, ha ha, roll coal. Well Cummings, Cummins? Cummings. They make oh man, so they make diesel engines for trucks, among other products, will pay a record Clean Air Act civil penalty of 1. 675 billion in a lawsuit brought by the U. S. Department of Justice, the EPA, and the California Air Resources Board.
Just like Volkswagen, they did the exact same software defeat process. It’s like in more than 630, 000 Ram 2500 and Ram 3500 pickup trucks. So hey, you want to get your diesel truck certified for you know, emissions testing? Well, our software knows it’s being tested and we’ll put on the clean zone. When you go out in the road, burn earth, burn.
Why are these maniacs doing this stuff? Why is it because the cost
Michael: I’m a little i’m at a loss somewhat on this one I mean, they they did this throughout the period when volkswagen was doing it then continued to do it even longer You know cummins isn’t admitting any guilt here, even though There is a software program embedded in the system.
They created that they essentially what it does is it fakes out the emissions test and emissions test is plugged in and the vehicle switches over to a different mode where it performs in a way that, that makes the emissions look all good and dandy. And there’s. And then as soon as you leave the the emissions test facility, the vehicle switches back into the mode.
I mean, there’s no way they didn’t know about that and design that. And they’re still not admitting guilt. Maybe that’s because they’re trying to protect some of their execs from going to jail like some of the Volkswagen execs did. But the fact that they’re saying, you know, they say they have seen no evidence that anyone acted in it.
bad faith and does not admit wrongdoing. I mean, that’s sheer bullshit. And you know, they’re, they’re essentially just paying off, paying their way or buying their way out of this one. I don’t like this at all.
Anthony: Yeah. So they have to pay not only a 1. 6 billion fine, which is a 1. 6 reasons of guilt admitting to guilt.
But then they also have to Repair and recall all of these engines, which, you know, we’re talking more than 600, 000 of that. So that’s got to cost a little bit more and I just wonder if, you know Is this literally the cost of doing business? Was it cheaper for them to do this than to create something correctly?
No, that doesn’t make any sense at all. This is just assholes assholing. Yeah.
Michael: There’s no way it’s going to be any cheaper now. I mean, I’m, I’m assuming though that the repair is going to be relatively simple software fix, so it’s not like the repairs are going to bankrupt them. Right.
Anthony: We commenting out a line.
Anyway how are we feeling about Tau Time? Tau Time. Tau Time this week, we’re going to cover a lot of things all at once. We’ve got a bunch of interesting articles together. One is about different, well basically the coverage is going to be about different alternatives. Oh no, not different, just alternatives to lithium in EVs for battery power and you know, Microsoft says they have, they have this cool AI program that does this massive search and they found an alternative to lithium.
And if you read the article, the alternative is lithium? There’s something called low batteries, which you can kind of fuel up your battery, which sounds kind of cool. But, you know, I’m just making things up. I don’t know what Mr. Perkins is going to mention, so go for it!
Fred: You’ve now entered the Well, thank you for the introduction.
It was something. So the, Article is the title, Unlocking a New Era of Scientific Discovery with AI. On Microsoft’s AI screened over 32 million candidates to find a better battery. And then they talk about how important this is because of the restrictions on procuring lithium and it’s scarce and environmentally destructive and all those good things.
As you read through the article, though, it turns out that these materials they selected after screening a half a million potential candidates all include lithium. So I think there’s a disconnect somewhere between their PR department and their engineering department. This is actually a fascinating article.
They talk about using AI to screen these different materials and the different analytic processes that support it. It turns out that it’s not the kind of AI that we’ve been discussing before, which involves a lot of machine learning by people sitting places saying pictures that, yes, that picture includes a cow, for example and this AI.
It’s really designed to simulate the benefits of quantum computing because it’s a really brute force approach using thousands of GPUs. Graphical processing units dedicated to this task to sort through a lot of information using rules that are being generated by human beings to filter the information.
So it’s, yes, it’s kind of AI, but AI doesn’t have any official definition, so they can stretch things a lot. So this is, this is really a lot more about. The potential for high speed computing and like so many other articles about AI, it mixes the aspirational with the actual and very clever ways to say that, well, we can do this today.
That means that in the future, all these other wonderful things are going to happen. It’s it’s actually a really interesting article because what they’ve done is they’ve gone in and they’ve simulated the actual electron densities of individual atoms in small samples of crystal structure to look at their chemical properties and how these can be used to improve The electrolytes that are in batteries in particular, solid state electrolytes that could be used in future batteries.
And they go through the process and talk about all the different things they do. And I don’t understand all of those things in detail, of course. I don’t have a PhD in chemistry. But it’s really neat work that they did. It’s just that it’s packaged in such a way that it’s really misleading, I think. I also note that two of the authors of the paper they reference are, in fact, Microsoft employees, which may have given it a bit of a bias.
Still, the work is really good. They’ve done, you know, they started off using a hundred, you know, half a million potential candidates. Sorted through them in different ways, and it came up with one that they’re actually building in the laboratory. So far, they’ve got some powder that they’re testing to determine what the chemical properties are and what the electrochemical properties are, which is even more important to them.
But it’s interesting that it’s, it’s also misleading. I don’t know why they are having so much urgency to show the value of high speed computing.
Anthony: Anyway. Go ahead. It’s all about selling stuff. Come on, check out our cool stuff. Well, yeah. I just don’t read the fine point.
Fred: I’m trying to sag it to current, popular items, I guess, in the press.
But anyway, so in the future, everything will be better, and some of these materials that they processed may, in fact, have some tremendous benefits, but we’ll wait and see. It’s a demonstration project to show that it could happen. Interesting, though, that because of the limitations on the study, which are significant, There may well be a lot of other candidates that they tossed out due to their screening process because they do not have a big base of chemicals that they can use to train the AI to pick out all the potential strong candidates.
So they’re really imposing a human bias on it with the filters that they put in. Very complicated calculations underlie this. So it’s, it’s, it’s a really nice project. But You will not, as a result of this, next year be able to buy a car with a solid electrolyte that’s based on this process. They make an estimate that they’ve speeded up chemistry by, what, a factor of 2, 500 or something like that.
It’s a combination of scientific expertise and AI. I’m quoting now, the combination of scientific expertise and AI that will compress the next 250 years of chemistry and material science innovation into the next 25. Well, that’s quite a claim. I don’t think that’s true. That’s kind of like saying that the invention of the slide rule will improve mathematics by a factor of a hundred, maybe, but not too many people using slide rules anymore.
So interesting article. I encourage people to read it, but take it with a grain of salt because the people who wrote the article are not the people who did the research. And there’s a big disconnect there, in my opinion.
Michael: Yeah, it seems like, you know, when I read the article initially, I was thinking that, you know, we’re getting rid of lithium.
We’ve got a way to get rid of lithium and batteries and some of the drawbacks it has, but after listening to you, Fred, not so much.
Fred: Well, the final material they came up with has what the, has both lithium and sodium in it. So it has two ways Of allowing ions to migrate through the electrolyte rather than just one, which is the lithium ion approach.
Right? So you’ve heard talk about sodium batteries. You’ve heard talk about lithium batteries. You heard talk about sulfur batteries, what they’re trying to do here is say, okay, if one is good, two has got to be better. So the material they come up with has both lithium and sodium ion transfer channels in them.
Hopefully that’ll make the batteries much better, but they’re a very long way from producing even a first prototype battery that uses this candidate material.
Anthony: So from an auto safety perspective, just remind listeners or educate them on what is the problem with lithium?
Fred: Oh, well, from the auto safety perspective, the problem with lithium is that it’s used in a electrolyte matrix, which is liquid and is also flammable.
And so when there is either a manufacturing defect. In the battery itself, or there is a mechanical damage to the battery case, you can have a self igniting fire that does not rely upon oxygen to burn, so it’s kind of like rocket fuel makes it very difficult to put out and also is creates a lot of toxic gases due to the Partial and uncontrolled burning of the electrolyte material very hot fires.
And of course, that sets fire to other things in your car and sets fire to your house and, you know, all those kinds of things. So those are all bad. Now, the numbers of. Cars that have catastrophic battery fires, as far as we’ve seen, are not significantly higher than the number of cars or gasoline engines that have catastrophic fires, but still the hazard is there that the fires can start spontaneously, they can reignite days after the original fire burns out, so.
There are those problems associated with lithium ion batteries, and people are finding out that if they’re not really carefully made, as in the case of low cost electric scooters they have a tendency to burn, and many fires in New York City. Have been caused by this. So those are the problems associated with lithium as well as the industrial problems of just getting the lithium out of the ground and all the, you know, the environmental degradation associated with extracting lithium.
So it’s, it’s not it’s not a warm and fuzzy material, right?
Anthony: Okay, so, we also have another article from IEEE about low batteries and this seems like it’s, you know, it’s still your electric car, but instead of plugging it into charging it, the idea is that you’d actually go to something similar to a gas pump and you’d Or in nanoparticles or something like that to charge your battery?
Fred: Nanofluid. Yeah, so this is a pretty nifty technology actually. So batteries run down and when batteries run down, they deplete the electric charges in the electrolyte. That’s what causes the batteries to run down. So, in this technology, rather than recharging the battery, they replace the electrolyte.
With new electrolyte. So they, the old stuff out and pump the new stuff in and off you go. And it could have a lot of, it could have a lot of benefits. But like hydrogen, for example, there’s no infrastructure available now for distributing this stuff. There are no cars that, that use this stuff. And I think that the authors of this article have glossed over a lot of the environmental problems.
They’ve said, for example, if it spills on the ground, you just sweep it up and away you go. But a problem with nanoparticles is that they’re really small. They’re less than a micrometer in diameter. Micrometer is one millionth of a meter. So it’s, there are a thousand microns in a in a millimeter. Just to give you an idea.
It’s really small. It’s like the size of bacteria. Because the nanoparticles are really small, they can slip in between cell walls. They can permeate through cells. They can lodge in your lungs. There’s a lot of things that happen with nanoparticles. In fact, the diesel emissions of nanoparticles are one of the primary reasons for California.
To push hard on getting electric trucks in the Long Beach area to try to get that pollution source with, again, nanoparticles and sub micron particles away from the people in that neighborhood who are suffering a lot of asthma and cancer problems because of that. So, there’s a long way to go in this. It is, it is promising.
I’d like to Hydrogen, for example, it requires a completely new distribution system and the vehicle’s delivery systems to handle this new fuel. So considerable hurdles yet to come, but it is a neat technology, replacing electrolyte instead of recharging the batteries and good, good, clean, fun.
Chemists are busy at work and good stuff will come out of this.
Anthony: So we can just pour Gatorade into my car in the future is what you’re telling me? Just dump the old Gatorade out, put the new Gatorade in? What
Fred: did what did Doc say? He needed cans to put into the, into the car, back to the future, I can’t remember his exact words, but yeah, something like that.
In the future, it’ll
Anthony: be better. Yeah, no, he had a fusion master and he was throwing like banana peels and stuff. Yeah. Yeah. Yeah. That’s a great one.
Fred: Hey. The future came and went and we still don’t have
Anthony: that. Hey, hey listeners, if you enjoy this, consider donating, go to autosafety. org and click on donate.
Donate, donate, donate. And with that, let’s jump into our recall roundup. How’s that sound? That sounds great. Alright, so this Didn’t we
Fred: that roundup in a piggly wiggly parking lot? Would that
Anthony: work? I Ooh, roundup. You know, I I I admit it That’s not the kind of roundup. Is it? Only if we buy product from Monsanto.
Um, I, I still haven’t been to a Piggly Wiggly. I, I want to, you know, I, I once took a class on retail and we talked about Piggly Wiggly.
Michael: Well, I mean, you’re going to have to leave the, the frigid north to even encounter one. I don’t think I start seeing them until, I want to say Alabama, maybe the Carolinas, somewhere in there on my, on my journey south.
Anthony: Yeah, well, the thing is, I don’t have to leave the frigid north because soon it’ll be the tropical New York. Come to the beautiful Caribbean beaches of Manhattan. I were talking nonsense now, mainly just me. Alright, recall roundup. We’re gonna start off with Volvo a rare entrant into the recall roundup.
17, 409 vehicles. This is the 2024 Volvo XC40. It is a hybrid electric. This is a software issue. They’re the, the software failed to warn an approaching driver their intentions to turn or change travel lanes, increasing the risk of an accident. Wait. Yeah. That’s
Michael: called a broken, a broken turn signal.
Anthony: Cause I was like, this is, it’s repelling to let them know their intentions. This is, this is some high end stuff in this car. Just a broken turn signal.
Michael: Basically they, they got reports that just indicated issues with the left. It’s just the left rear turn indicator as well. So the left rear turn signal wasn’t working right.
They got four reports, no crashes or anything, but they went in and said, Hey, we need to do a fix here. And so it’s going to be a software update and it’s going to be an over the air release. They’re just not quite sure when it’s going to be coming to you. It says that they’re going to have new software entered.
Introduce before Christmas of 2023. So maybe they’re, they’ve already put it into new production. So maybe they’re getting that updated out pretty soon here for owners, but essentially it’s a left turn signal didn’t work and now they’re going to fix it.
Anthony: Or they should just be like UPS drivers and don’t make left turns, you know, refigure out your map.
Next one we have is from lucid. Wait, lucid. Come on. They’ve never recalled a vehicle, ever. Oh, wait, I’m kidding. They’ve recalled all of them multiple times. This is 2042 vehicles, the 2022 to 2023 Lucid Air, which I think costs a half a million, billion, trillion dollars failures in their early version of their high voltage coolant heaters supplied by wa Basto the HVC eight supplies warm air for both.
Cabin heat and defrost capabilities. The lack of defrost capability poses a risk. No one also poses a risk buying a Lucid. Wait, is that too hard? I, and you know, before Lucid, I was all gung ho for Lucid. I thought, this is pretty cool. I mean, I would never be able to afford one, but I liked their approach.
And it turns out it’s just been smoke and mirrors. Yeah.
Michael: This one’s, this one’s an issue because it’s, it looks like they are, what they’re doing here is, you know, they say, Oh, we’ve got this problem. We’re going to recall it. But they’re saying that essentially owners are going to have to have the failure occur.
Before they’re going to replace the component. So that’s kind of mind boggling to me. And I think it shows how new Lucid is to this industry and to NHTSA. That’s not something that typically happens with recalls, right? Recalls are a, a unreasonable risk to safety. And so a typical recall. Corrects the problem before it occurs here.
Lucid has constructed this and I’m not even sure if NITSA look at this one close enough before they put their rubber stamp of approval on this one. But lucid is saying that what’s going to happen is when the failure occurs, it’s going to send a message to lucid who’s going to tell the owner they need to contact lucid for a replacement versus just saying.
Hey, we’re going to replace all these and make sure this problem doesn’t occur. They’re saying, Hey, we’re going to wait for this problem to occur and then have the put the burden on the owner to reach out to us to replace the part. I don’t like this. I don’t like this at all. I don’t, if If, if, if a lot of manufacturers start trying to pursue this type of, of remedy, we will be screaming about it from the mountains, the hills, wherever they are, wherever, whatever makes our voice louder, because it’s just, it’s backwards.
You know, you need to be fixing a safety risk and addressing it before it happens and not before someone’s defroster is not working and they have a crash because they have poor visibility or run over, you know, a neighborhood child or dog. Fix the problem first. And this, this is just not a recall to me.
It’s, it’s a very, it’s kind of like a customer satisfaction program that’s favorable to the manufacturer prevents them from having to replace all of the crappy parts that they’ve put on cars until there’s an actual failure. So, and they know that some of their owners, well, lucid owners, maybe. Not less so than others, but they know some of their owners may not even complain about the problem or may not even take any action and that’s going to save them money as well.
So this is a really bad recall remedy. Not something that I think we ever want to see again.
Anthony: Yeah, I think this just comes down to Lucid saying we don’t have a lot of money left and we realized that most of these people bought this car and they’re just kind of like putting it on the equivalent of cement blocks in their front yard because we’ve recalled all of them twice.
Have they recalled all of the vehicles twice?
Michael: I know they have I know they did for some of the battery failure issues. They were having a lot of loss of power problems. Yeah, I mean, you
Anthony: don’t, it’s an electric vehicle. What a battery, who cares? Yeah.
Michael: But look, they’re funded by the Saudis, so they’ve got an endless source of oil coming out of the ground and being sold to fund their electric vehicle operations.
So I’m not too worried
Anthony: about that. When I buy a Lucid Air, does it come with a bone saw? Oh, is that too dark? Sorry. That’s
Michael: really dark. Yeah. Well, last one. Is that a Khashoggi reference? Yes. That’s a pretty good one. Yeah. Bone saw and acid.
Anthony: Yeah. Yeah. Yeah. Well, you
Fred: know. So you’re saying the Lucid owners may not be Lucid?
Anthony: Aha! There we go. They’re not getting enough air. Alright, let’s, enough of this nonsense. We got one last one. This is Alfa Romeo, right? They’re still my cars. This is an investigation. The Office of Defects Investigation has received 12 complaints and multiple field reports alleging incidents of stall, loss of motive power due to a fuel pump issue.
When does the investigation become a
Michael: recall? It’s generally dependent on the manufacturer. I mean, I think before most. Investigations are opened. You know, I don’t believe that’s a just opens them out of the blue without communicating with the manufacturer. I think they say, Hey, look, guys, we’re seeing a lot of incidents with this.
We’ve got a complaint here that, you know, we’ve seen a crash involved and, you know, we think that your low pressure fuel pump is failing on these vehicles and. Alpha Romeo comes back and says, well, we, we don’t really see a problem there. And it’s, it says, well, we’re going to open a recall here. So I think they give them at least some notice there because, you know, give them an opportunity to, to do the right thing and do a recall.
We, you know, here it’s, it’s, they’re, you know, they’re, they’re 12 incidents here out of 22, 000 vehicles, you know, that’s, you know, a fairly low rate. But it’s suggests that, that NHTSA has seen some reason to believe that these, these fuel pumps are going to continue to fail and continue to create safety problems on the road.
So I expect that. They’re going to, they haven’t, it’s just been open. So NITS has got a lot of work to do as far as collecting information from, from Chrysler, who, who built these vehicles. And then we’ll see how the investigation progresses over the next few months, maybe years.
Anthony: All right. Well, with that folks, make sure to get your recalls taken care of, wear your seat belts, and as always donate and donate again to the center.
Thanks for joining us. Bye bye.
Michael: Thank you. Bye bye. Bye everyone.
Fred: information, visit
Michael: www. autosafety. org.