Latest / Data Driven with Lou Peck / Rick Ruth: the evolution and future of EDRs, the industry’s global shift, and the Wild West of ADAS.
Transcript
- Lou (00:00:19): This episode is brought to you by Lightpoint, of which I'm the Principal Engineer. Lightpoint provides the collision reconstruction community with data and education to facilitate and elevate analyses. Our most popular product is our exemplar vehicle point clouds. If you've ever needed to track down an exemplar, you know it takes hours of searching for the perfect model awkward conversations with dealers, and usually some cash to grease the wheels. Then back at the office, it takes a couple more hours to stitch and clean the data, and that eats up manpower and adds a lot to the bottom line of your invoice. Save yourself the headache so you can spend more time on what really matters. The analysis. Lightpoint has already measured most vehicles with a top of the line scanner Leica's RTC360, so no one in the community has to do it again. The exemplar point cloud is delivered in PTS format, includes the interior, and is fully cleaned and ready to drop into your favorite program such as CloudCompare, 3ds Max, Rhino, Virtual CRASH, PC CRASH, among others. Head over to lightpointdata.com/datadriven to check out the database and receive 15% off your first order. That's lightpointdata.com/datadriven. (00:01:23): All right, my guest today is Rick Ruth. Rick is president of Ruth Consulting, which specializes in passenger vehicle event data recorders, EDRs, of which we'll be talking a lot about today, and restraint systems performance. He has published over 20 EDR related papers and assists civil attorneys and prosecutors in Frye and Daubert hearings to facilitate admission of EDR data in court. He teaches EDR imaging and data analysis to law enforcement and private reconstructionists via SAE and IPTM, and is a regular speaker at national and regional conferences. (00:01:55): Rick is actively involved in EDR committees for SAE, ISO, and ASDM. Prior to consulting, Rick put 33 years under his belt as an engineer for Ford Motor Company and, since 2008, managed the engineers who performed field investigations of safety system performance in real world crashes, including EDR imaging and analysis, and championed the release of Ford EDR data to the Bosch Crash Data Retrieval System. He was a member of Ford's EDR policy committee, was Ford's representative to the SAE EDR Standards Committee and helped shape Ford and Auto Alliance responses to the NHTSA on part 563 EDR legislation. Mr. Ruth has a BS in Electrical Engineering from Michigan Technological University, an MBA from the University of Michigan and is also a registered professional engineer. So that should give you an idea that Rick is the man to talk about this stuff with. So thanks so much for making the time to be here today. Rick (00:02:49): Happy to be here, Lou, anything we can do for help our fellow reconstructionists out. You're well known for it as well, so let's get on with it. Lou (00:02:58): Yeah, that's what I was going to say. You have a reputation for being available and I think everybody really appreciates that, especially with your unique background, your unique perspective, and part of that was your background at Ford. So I saw that you started there in 1973. I don't want to rub it in or anything, but this is eight years before I was born, and then you were there to 2006. So when did EDRs even appear on the radar and what did that evolution look like while you were at Ford? Rick (00:03:29): Sure. Well, if we really go back to the beginning, I was hired to do the airbag firing system in 1973. The technology wasn't ready yet. I would say have a hard time falling asleep and wake up in the middle of the night fearing that people were going to die because my system wasn't going to work. There were mechanical systems and the old mechanical systems was contact, bounce and other things that kept things from working properly. But when EDRs came in was when we went to electronic sensing systems. So GM started at '94 and Ford said, "I'm not so sure about this, let's give it a few years." So we gave it until '97 and then we began with electronic sensing and began to put in very small data recorders for just the acceleration data that was used to make the decision to fire the airbags. (00:04:26): Now, we were a little bit jealous of GM because they put speed data into theirs and they covered a longer time period, but we had an attorney, a Vice President of Privacy, and he said, "If you don't need it to exactly understand how your system worked, you can't record it. We don't want to violate people's privacy." Well, the great news was we downsized that vice president in 2002, and then when the cat's away, the mice will play. So Ford engineers said, "Well say, what should we do?" And say, "Well, you don't have to reinvent everything. What's GM doing? Okay, all right, let's do something like that, but we'll have to change it just a little. They record throttle angle, nobody wants to record throttle. We want to know what the driver's doing, we'll record accelerator pedal positions." So anyway, we made the Ford EDRs and say it came along, but we had a little bit of a slower start than our friends at General Motors did. Lou (00:05:22): And so when was that first EDR, that most simplistic EDR that was capturing acceleration profiles? Rick (00:05:29): For Ford 1997 for GM 1994. Lou (00:05:32): Okay. Yeah, I think I had a 1994 Pontiac Grand Dam, and I'm pretty sure that had an event data recorder on it. Fortunately, I never tested that out even though I was 16. I somehow avoided crashing that vehicle. So now we're at a spot where 99% of the vehicles rolling off the assembly line have an EDR. Is that right? Rick (00:05:54): Yes, and the great news is that we actually have to thank the United States government for that, because they made a regulation, well, technically that regulation didn't require vehicles to have an EDR. For all practical purposes, the government was sending a message of, "You really ought to have one of these, and we know all of you're recording something. If you record anything, you got to record a little bit more, make it a really good one." And because we have that foundation that was laid in the United States 10 years ago that we have event data recorders in the vast majority of our vehicles, certainly almost all the new ones. And because we've been putting it in all the new ones for over 10 years, the number of old ones and new ones, the percentage that we have is in at least the middle sixties approaching 70% of all vehicles on the road in the United States that have an EDR. Lou (00:06:52): Wow. Yeah, that's incredible. So it's something to the effect of, I mean if you do the math there, you're going to have 80% to 90% of crashes, at least one of the vehicles is probably, statistically speaking, going to have some sort of EDR on there. And it's been a long time since I've taken engineering stats, so that might not be right, but that sounds right. And early on, I think you probably remember, and I started on the scene in the early 2000s, there were certain reconstructionists who said, "Listen, this is where I'm drawing the line. There's no way I can keep up with all this. I'll hire somebody to do the EDR stuff. I'll hire somebody to interpret it." And that just kind of proved to be a methodology that's not really feasible. You have to at least have a solid foundation it seems, in EDR now if you want to be a relevant reconstructionist. How do you think of that? What kind of level of knowledge is required to be a reconstructionist nowadays on the EDR front? Rick (00:07:48): Sure. Everybody doesn't have to be an expert, but the way that I think of it is if you show me all the scene evidence, I can solve about 80% of the crashes with pretty good accuracy. If you give me the EDR data, I can solve about 80% of the crashes with pretty good accuracy. But if you want to get them all right, you got to fit that physical evidence and the EDR together. And so that's why the reconstructionists are typically not farming out the fundamental EDR analysis that everybody needs to do it. Now, occasionally you come up with a really weird one, a really unusual one, and that's what I do all day every day, is field the calls from the people that have weird ones and say, " I'm pretty good at EDRs, but I just can't figure this one out. It's driving me crazy. Can you help me?" (00:08:37): I'll go through and look at it, and they'll say, "Well, I think your car hit the curb here, and then the wheels bounced up and the wheels spun when it got in the air. And so you have a crazy stupid speed reading and it's confusing you." But if we look at everything else and figure out where was the EDR when it wrote down each number in it, then we can make sense out of it all. So the secret is fitting the two together. And so while it helps to have a specialist like me available for the weird ones, the typical reconstructionist basically has to be able to look at a report and get the fundamentals out of it and say, "Do the fundamentals match my scene evidence? Does it match the rest of my traditional reconstruction, my momentum, my crash, my time distance for the things that lay out there?" So I think everybody's got to be this good at it, and a few of us have to be that good at it. Lou (00:09:32): And I always appreciate the fact that you're available for things like that. And I've had several cases where the EDR data all makes sense except for polarity seems to be reversed. There's something like that. And before I'm comfortable moving to deposition or trial, I just want to reach out to somebody who has their finger on the pulse of a lot of these anomalies and is continually in touch with the community. So, like you said, we need people like that in the industry, and I'm glad you're there and I appreciate your willingness to share. Rick (00:10:05): And just a quick note, Lou, that actually the United States government could be putting me out of business because of their wonderful Crash Investigation Sampling System, which now has over 14,000 crashes in it, of which 8,000 have EDRs. If you now have something like an anomaly of polarity, you can go in and you say, "Well, I have a 2014 Nissan Altima with this polarity. Let's go look at the other 2014 Nissan Altimas." And you say, "Oh, well in this one the airbag's deployed, but it said I got hit from behind." You'd say, "Oh, I think we have a polarity problem," so you don't have to call me all the time. Actually, you can look in the CISS database and figure out a lot of that stuff, which is very helpful for our community. Lou (00:10:53): Yeah, it really is. That's huge. And there's a link to that on your website, ruthconsulting.com, and you go under vehicle exterior CDR maybe. Where does it go from there? And then you can actually download the CDRX file from any case? Rick (00:11:07): Yes. So you were describing the chain. So, as you say, once you get into the event, it's click on vehicle exterior, and then there's an EDR, and then you have to click the Expand button from EDR, and then it has Summary and then Event Record One, Event Record Two, you click on the Summary and that's where the link to the CDRX file is, which is, because they're small files, very easily downloadable. Lou (00:11:33): And you just have to figure out, you could be tricked into thinking that file isn't there. So they buried it on us. But that's a great point. That's available and that's huge for looking at, and then looking at similar damages that you might be seeing in a case, and now you actually have a documented Delta V or acceleration profile from that. That's awesome. And rumor has it, Rick, they're going to start doing motorcycle crashes soon too, so I'm excited about that. Rick (00:12:00): Well, that's fabulous because I was going to say, because they not only have the EDR report, they have the scene diagram and they have photographs of the scene, unfortunately not with cars at rest in them because they get there three days later. But then they do have photographs of both vehicles taken in the storage lot, and in some cases they're three, four, five, six vehicles in some of them, but they have a lot of photographs. So when you can have a damaged photograph, a scene diagram and an EDR report, you should be able to figure out what happened on that one and to see if it's similar to your crash. And that's what we're all looking for is something similar to our crash that we can say, "Okay, here's how the EDR behaved in this one. So that's how it should have behaved in mine and say, figure it out." Lou (00:12:48): Yeah, that's huge. They do a really good job over there. So we touched on a little bit already, but how has the data evolved since you started? So at the beginning it's literally just capturing, from what I can understand, the acceleration profile of the crash, and then you could integrate that and get the Delta-v. And now we've got GM's Advanced Safety Control Module. So what's that evolution look like as far as where you started and where we're at today? Rick (00:13:16): Well, you got to start perfectly right. Ford happened to record acceleration data. GM recorded Delta-v once every 10 milliseconds. So Ford ran out of memory before the crash was over. Lou (00:13:27): Not great. Rick (00:13:30): Yeah, great detail for the beginning of the crash where the airbag deployed, which is what Ford cared about. Whereas General Motors took the longer view of saying, "No, I want to see what the whole crash looked like, and I'll accept a little bit less detail right around where the airbag deployed." But that was the beginning for both companies, was just what went into the inputs of setting the airbag system off. I think really the biggest evolution along the way was when NHTSA got involved planning for the regulation in 2006 is they published, their intent was to change the intent of the recorder from being for the manufacturer, knowing if their safety system worked correctly, to being used for effective crash reconstruction. (00:14:22): And that's the key thing, is who's the customer for this? Is we change the customer from the auto company to the reconstructionist, so we should all be thankful they're doing it for us, they're doing it for us now. So that was their intent and that's when we began to see the speed data being added in. And then I was really proud of Chrysler. They put in the stability control system data upfront, and that wasn't regulated or required in any way. And similarly, when I was in Ford, one of my last dying acts before I retired was we had the big meeting where we said, "Everybody's got electronic modules, everybody would like to store data, but we really should think about putting this all in one place all together and making one EDR that can serve many masters in terms of knowing that the systems performed properly." (00:15:16): So you had people out there that ran the anti-lock brake system and others that worked on the tire pressure monitoring system, and others that were worried about engine fault codes and say, "Well, what if we could take all that and put it into one?" So I was part of those meetings that created the 2009 Ford system, eventually, and it was a pretty good all in one recorder for 2009, and it was in fact far ahead of what the government was asking for. And then of course the government regulation took effect and there was a minimum floor that was set for all manufacturers. But I think what you really noticed was that there was a difference in the spirit of the manufacturers. So GM, Ford, Chrysler, Toyota, all took the tack of, "I'm going to record as much data as I can and share as much data as I can." (00:16:13): Whereas some of the other manufacturers, some of them had either European influence or Asian influence that said, "No, owner privacy is important. We're not going to record things that aren't absolutely necessary." And so you saw the Mazdas and the Mercedes that did the bare minimum that was required by regulation and did not go beyond that to providing extra helpful data to we crash reconstruction. So they did what they had to do but didn't go beyond that. So I think what you see is the big four, GM, Ford, Chrysler, Toyota, in the United States anyway have all taken the attitude that more data is good. And so, along the way, you've seen some major points where the amount of data has gone up. So certainly every generation of GM has included more data, and their most recent one, the SDM 50 and the 2021 Corvette is incredible in terms of what it has. (00:17:16): And GM has added that Advanced Safety Control module and the Front Camera module. So GM, they started as the leaders, they got behind for a little while, behind Ford and Chrysler. Now they're way back out front again in terms of the amount of data that's being provided. So if you say what has the history been, it's just been more and more data every year. Not necessarily a tiny increment every year, but a major leap forward every three, four, five years by different manufacturers. And then if you take the aggregate of it all, it's been a little bit more every year for us, the reconstructionists, but we get it in lumps when we finally get it. Lou (00:17:59): Yeah, that ASCM, I think you and I were both at IPTM last year, 2022, and I sat through Don Floyd's presentation, which was awesome, and he seems to be ... Don Floyd, for those that don't know at GM, the way I understand Don's role is he's heading up their EDR efforts and pushing the boundaries. And like you said, GM has kind of always been doing that and is taking the lead again. And hopefully we get some other people that say, "Hey, that looks nice. It'd be good to have that info." Rick (00:18:29): Yeah, Don is a global, he used to be more like the United States EDR guy, and he's now the global EDR guy for General Motors. So he's also sucked in deeply into the China issues and European issues, and United Nations issues. And it's like herding cats for poor Don. Lou (00:18:51): Yeah, and we'll get into what's going on in the world market in a little bit. Hopefully that doesn't pull them away too far from continuing to push the envelope and get more data. As greedy reconstructionists, we want as much data as we can get. So the best tool for us obviously right now is the Bosch, the CANplus, and then the CDR 900. We need them both to do our job at this point it seems. Rick (00:19:15): Absolutely, there are still many vehicles that you can only use the CANplus on, some that you can use either interface on, and then a small but growing number where you must use the CDR 900. So you absolutely have to have both tools in your bag. Lou (00:19:32): Yeah. And that's where we're at. And how have you been making out with the CDR 900? I know just at random downloads, everybody kind of gripes about communication issues, but I feel like you've probably solved that issue for yourself already. Rick (00:19:45): You'd like to think so. I have trouble too. Lou (00:19:50): Okay. It's good to know we're not alone. Yeah. So yeah, I saw Kent Boots recently posted up on LinkedIn and he said, "Hey, make sure the module goes beep before you plug it into the USB port on your laptop." And I was like, "All right, maybe that's all it was. Maybe it was just something as simple as waiting for the beep." Rick (00:20:07): I think it's wait for the beep before you try to communicate with you. Actually when you plug it in, it gets the 5 Volts and then it needs about 10 seconds to give you the beep that it has 5 Volts now. It's still not ready to do its whole job. You need the 12 Volts plugged in to do the whole job, but I think you'll get the beep even when you just plug in the USB. Lou (00:20:27): Yeah, the way that he was mentioning it for, let's say, a normal DLC download is plug in into the OBD two port. You'll get the power, wait for the beep and then apply the USB cable at that point. So hopefully that does it. Rick (00:20:41): I've gone ahead and put the USB in first, and it's been working for me lately but, in the early days, I do remember spending say three days at eight hours a day with the Bosch helpline, and after about 24 hours of sitting on the end, they said, "By the way, is your VCI Manager still open?" I said, "Yeah." He said, "Oh, well it won't work if you do that." Lou (00:21:04): It was as simple as that. Rick (00:21:06): It's like, "Could we have said that 24 hours ago?" Lou (00:21:09): Yeah, exactly. And they're going through some big changes, you and I were talking about before we were recording, they've, historically, the R&D, Bill Rose, they've been in Santa Barbara, and now they're spreading, they have global efforts now they're spreading out their R&D center and they're having problems with supplies and getting components. You try to go buy a CANplus unit right now. I don't think you can. So what's going on with Bosch as of recently? Rick (00:21:38): Well, I think you're seeing a transition from Bosch being focused on meeting the United States regulation and the needs of the manufacturers to meet the US law to now saying, "Wait a minute, we've got China with an EDR rate. We've got the European Union that has a regulation that is now adopting the United Nations standard. So clearly this is going to be a global operation, and so we shouldn't be focused only on the US." And we know for cost reasons that many people are saying that India is a lower cost source of engineers, especially software engineers. And so while I hadn't heard an announcement of it, but by speaking with people in the industry, I've become aware that they have basically downsized the United States staff except for the Program Manager, Bill Rose, and they've gone and sent most of the development and the testing to India. (00:22:40): And of course I'm sure they're having to staff up in India, but anytime that you lose some experienced talent as they have with their United States people and you're trying to gear up, say people to become familiar with a system over in India, there's a learning curve that's going to be in there. And so I think we're in for, call it a rough year or two until the Indian operation is fully trained and fully staffed. Now, those are just my thoughts. I'm not speaking anything from Bosch internal. I haven't heard any complaints from Bill Rose, but I suspect that he's probably pulling his hair out a little bit nowadays from my own experience with trying to use the Indian culture to do things that they are not yet familiar with. I think they're fine people, but I think you take anybody and you give them a job that they've never had before and have to understand a complex global system. I think there's going to be a learning curve involved there. And I think we're going to see some bumps along the way from Bosch until that staff has more time to learn and implement EDRs globally. Lou (00:23:57): They have so much to deal with. They're taking feedback from us in the field using the tool and saying, "Hey, we had this weird issue that we've never had before. Can you create a patch to interacting with the manufacturers, which of course you're intimately familiar with, and implementing any changes they're making or integrating new models, like the new Corvette?" Goodness gracious, that's a lot to deal with. And then, like you said, you just put a new group of people together on it and it does sound challenging. So hopefully, the growing pains don't affect things too much and they smooth on through it. And then supplies, like you were saying, can't go get any cable you want right now. So if I find out next week I have to go download a brand new Corvette and I'm expecting to go direct to module, the odds of me being able to buy that cable, who knows? Maybe I can, maybe I can't. So to have relationships with other people in the community, like you, you could bring them up and say, "Hey, do you have that? I'd love to rent it from you." Rick (00:24:54): Yeah, I literally just sent a Subaru cable to a colleague last night because they couldn't buy it. It was not for sale. Now, while cables don't have chips in them, what we do want to say in terms of things like the interface modules is there still is a very serious global chip shortage going on, and that's affecting not only things like interface modules, but EDRs. The EDRs themselves, GM I'm sure would love to have their new SDM 50 in many of their models but can't get the chips. So the combination, and it's not just airbag modules, it's every electronics module in the vehicle, any analog brake system, engine controllers, the global chip shortage is having an incredible effect on the overall industry in terms of the ability to move things forward and get all the parts you need for the design. So certainly Bosch is just a small part of that. We're seeing supply chain issues in general, but I just highlight the chips in particular as being a known ongoing issue. But supply chain, even cables are hard to come by. Lou (00:26:06): That makes it tough. So with the Bosch system, we can tackle the majority of the downloads, but there are a couple ancillary systems that either you need to own or somebody in the community might need to own to interrogate a system. So what are the secondary systems nowadays? Rick (00:26:23): Well, the biggest one is Kia and Hyundai. So they made their decision when the EDR regulation came out that instead of using Bosch, that they would go to their dealer diagnostic toolmaker, GIT, Global Information Technologies, and say, "Well, GIT already knows how to talk to all of my electronic modules. Why would I want to pay Bosch who doesn't know my modules, how to go learn how to read them, when I can just tell GIT, my captive diagnostic toolmaker, thou shalt make this tool for me?" Just as GM did the Victronic back in 2000, and initially Vitronic said, "Why would I ever want to do that? That doesn't sound profitable." GM twisted their arm and said, "How many tech twos do I buy from you every year? You will make this tool for me." So I think GIT was placed in a similar situation where they were told they were going to make a tool. So it was unusual that Kia and Hyundai, while they are tied together back in Korea, that in the United States, I believe due to some say little known antitrust issue, that somehow they weren't allowed to talk to each other in the United States. That Kia and Hyundai headquarters are five miles apart from each other out in Irvine on the 405, the 5, one of those? Lou (00:27:52): 5 probably, yeah. Rick (00:27:54): Okay. But they're not allowed to talk to each other. Lou (00:27:57): You put the "the" in front of it. That's very impressive. That's a SoCal thing to put the 'the' in front of the 5. You're clearly a cultured and a well-traveled man. Rick (00:28:05): Didn't know that. But technically there's this wall between them. So you buy this Kia tool or you buy a Hyundai tool. Anyway, they ended up making two tools, whereas the interface module functions identically. So the only difference is they put blue plastic on the Hyundai ones and red plastic on the Kia ones. Other than that, they are functionally identical. So when I say we didn't need to buy both tools, they made it so you had to buy both tools because you could only update the software, load and update the software, for the one you bought the tool for and had the license for. So I just went and bought both licenses and put both softwares on my laptop. Lou (00:28:50): Yeah, that's smart. Well, I think it's nine grand right now from the Crash Data Group to get both of those. The price that I don't know, and maybe you do, is how much is that annual upkeep on the GIT stuff? Rick (00:29:01): It's $695 per brand, per year. Lou (00:29:03): ...on the GIT stuff. Rick (00:29:03): It's $695 per brand per year, currently, so for the two of them, $1,290, which is pretty close to what Bosch is for the CDR system. Bosch is $1,250 last I looked. Lou (00:29:15): Yeah. It's a hard pill to swallow when it's just doing Kia and Hyundai and Bosch does everything. It would be great if they go the path of Subaru and eventually get over to Bosch, but I don't think that's on the horizon. Rick (00:29:27): I don't think it's going to happen. Lou (00:29:28): Okay. We talked a little bit about Part 563 and how the federal government said, hey, if you are recording stuff, this is how you have to do it at the bare minimum. Then last summer, they came out with this notice of proposed rulemaking and they said, all right, you got five seconds now of pre-impact data at two hertz, two samples a second. We want you to do 20 seconds of pre-impact data at 10 hertz. That was the biggest part of the rulemaking that I saw and apparently, I learned last night from you that was shot down, huh? Rick (00:30:01): Well, when I say it shot down the manufacturers, they're all given 90 days to respond to it and 201, they uniformly responded, that's stupid. Don't do that. By the way, your cost estimate of that it would cost 0.3 cents per module to do that was way off because the government doesn't know how to do cost estimates very well. They don't understand that. They think it's just making a little bigger memory chip. They don't understand all the processor overhead and the RAM memory that has to be buffered to feed that little bit of double EPRA memory. But the government missed a huge opportunity. The reason that they went after the recording time and frequency was that when the FAST Act was passed back in say 20, say '15, '16, '17, the government basically said, we don't think the timing is right. Somebody should go study this and come up with what the right amount of time is. (00:31:06): The way the government works is they let a contract of Virginia Tech. Virginia Tech went out and collected a group of statistics and in the end, from what I can see, they after they collected all these statistics, ignored them all and just said that it should be 20 seconds at 10 times a second. There really wasn't a statistical rationale for it. When I say that government missed a huge opportunity, they basically did what Congress told them to. Go look at the time. Well, the original regulation was formulated in 2006. (00:31:41): In 2006, stability control was not yet, say, something that was required. That phased in as a requirement from 2009 to 2012, so stability control system data is a big thing to us in the crash reconstruction world. It was not really included in the original Part 563 regulation. Now, if I was going to update Part 563 and I was on, say, the consulting committee, I'd say we want that stability control data. We want want all of that stability control data, the yaw rate, the steering, the lateral acceleration, longitudinal deceleration on the zero to one g scale so we can see more precisely how fast a car is speeding up or slowing down. (00:32:29): That is something that's been around, I should say, been in cars since 2012. Well, then we came along with automatic emergency braking and that has gradually been phasing in. There was a voluntary agreement. NHTSA decided to not regulate automatic emergency braking. They decided to make it a voluntary commitment by the industry. At the earliest time when they asked for commitments, it was for generally, how about by 2022? You see a lot of that. Not everybody has made the 2022, but the industry is pretty well along. But again, along with automatic emergency braking came additional data, but more specifically, is the car breaking for you or not? Is there a target in sight? How far away is that target? At what point does it become a threat that needs to be responded to? And then did we warn the driver that there was a threat? And then did the driver respond to whether or not there was a threat? If the driver didn't respond, did we overrule the driver and decide we were going to stop the car anyway? How hard and how fast? Lou (00:33:38): Yeah. That sounds very valuable. That's something I would like to know. Rick (00:33:42): I think we'd all like to know that. Now, General Motors has been kind enough to share that with us. Toyota has shared a little bit. Honda has shared, are we warning, are we braking in their EDR. But basically, there's a ton of data and we say, so that the government made this new NPRM, notice of proposed rulemaking, and what do they focus on? Time. There was so much more. I think if you read the industry responses to the proposed regulation, industry themselves were saying, hey, you're missing the boat here. (00:34:21): You don't get it. Time is not the problem. Say the problem is we've got all this extra data and some of us are sharing it voluntarily, but if you want to advance the state of the art, you've got to go from voluntarily to if you have it, you got to write it down. That's been the way that many of the other regulations around the world have been written. They're saying, we recognize you may not have it in every vehicle, so you don't have to write down something that you don't have the sensor for, but if you have the sensor, you got to write it down in the EDR. Lou (00:35:03): Not only, obviously, would it help us as reconstructionists, but if we're trying to evaluate how effective these autonomous systems are, then we need some way of judging them. If we don't know if they braked or not, then how are we evaluating their efficacy? It's tough. Rick (00:35:21): Yeah, no. I think, say, we may be skipping ahead just a little bit there that when you say how would we know? We don't have a data collection system set up just yet to collect. Of course, if the collision avoidance system is successful and avoids the crash, no harm no foul. No one goes to look at the data that says, good job. We avoided that crash. The only data we typically get to see is the ones where it has failed and the car did actually crash. Then we go look at the data and then we get to see when was the warning issued? When was the driver response? When was the automatic response? (00:36:02): But basically, I don't want to characterize it too harshly. I don't want to say it was a system failure because many of the systems, as we are developing and evolving automatic, say, emergency braking systems, we've evolved from just looking for a car that's moving on the road straight ahead of us to looking for a stopped car on the road ahead of us to looking for cars intruding from the sides to cars coming in intersections to pedestrians walking, say, across the road sideways, to pedestrians walking along the road front ways, to pedestrians along side streets of streets we're crossing and then there's the in bright light, in dim light, in darkness. (00:36:48): We have a huge evolution that's taking place. When I say when there's a crash, there's a failure of the system, I don't actually mean a true failure in the sense that it didn't do what it was designed to do, but if we say you take a 2016 crash avoidance system, it was designed to do this much to look for cars straight ahead. You look at a 2022, 2023 system, and they're looking at a much wider set of variables. There were many collisions that weren't designed to be avoided in 2016, '17, '18, '19 that are now being designed to be avoided. But you say, so what are the ones we studied? We studied the ones where the system did not yet avoid that collision and that has helped us to make systems which can avoid that collision in the future. Lou (00:37:43): Yeah, that's huge. Tesla is doing things like fleet learning, where they're sending a bunch of data to the mothership and processing it with machine learning and trying to figure out how to make sure that it doesn't happen again. I'm sure other manufacturers are doing something similar, whether or not that's their methodology. They're not necessarily sending it to the cloud, but they're analyzing the data, they're analyzing the crashes. I have to suspect some manufacturers are going to that CISS system and pull in the data and looking at how their system performed. When we go look at the data, you touched on it a bit already, but who does give us data in the CDR report with respect to ADAS function? Who doesn't? And then when they don't, are we getting it places like vehicle control history or Tesla's vehicle data report? How does that look? Where do we get ADAS performance data and who gives it to us? Rick (00:38:33): Great question, and I should have had a list made for you, but I don't. We of course know General Motors is giving us the advanced safety control module, so they're the forefront of giving us data. We can say Ford, to my knowledge, does not yet, to my knowledge, although I have to be careful. The latest RC8, I need to double check to see if it's in there or not. Say Chrysler has begun in some of their vehicles to say, we're warning or we're braking. Toyota, I don't think has put it in the CDR report, but they clearly have it in the Toyota Techstream data that we can access publicly. I've noticed in Honda, they started putting it in gen twos and 2016. It has it, is the system reacting and if so, at what level that it is? I haven't seen it in all the Kia Hyundais, but starting with the 2019 Forte and in some of the more recent models, I've seen it showing up in Kia Hyundais. I don't have a great feel for it. I haven't made that chart yet, but now that you've asked, I think I will. Lou (00:39:46): Yeah, and that's another thing I meant to mention. On your site, your site is awesome. I find myself there with some frequency. You have a lot of charts and compatibility, what vehicles are supported and what they offer for data, so we really appreciate that. It's clear, if you have to make the chart, then we're in the Wild, Wild West past Part 563. If you have an EDR, Part 563 governs that. But anything else, it's just up to you and it's up to the reconstructionist to figure out what might be available via the various tools. Rick (00:40:21): Yeah. When I teach the analysis classes, well, I try to give people a good understanding of what data is currently available. Our mantra is just go get the EDR data, open the report and look at it, and see how much have you got to work with? We can make all the charts we want to predict what data you will get when you get there, but the bottom line is get there, get the data, just open it up, and see what you got and then use everything you can to the best of your ability. Lou (00:40:53): Yeah, and GM is killing it. Like we were talking about earlier, we got GPS data now. We have a clock. We get information about the target vehicle and the closing speed there. And then, like we were talking about before, ADAS intervention. Do you think anybody will follow suit there? It's obviously great to have the GPS data. We know where it happened, we know when it happened. That bridges one of the big gaps in recon, which is, well, you got to sometimes do the whole recon just to figure out if that's your data. Do you think people will follow suit? Rick (00:41:26): I think you're going to see the Wild West still in progress and I think you're going to see different manufacturers making different choices. One of the analogies I like to use is at the time that I left Ford, it was a dark day for the automobile industry, but basically, GM and Chrysler went out of business and Ford came that close. But the CEO was laser-focused and the CEO came out to all the staff and said, listen, we're getting rid of half of you. Of the half that are left, you can't do everything you did before, so here's your guidance. (00:42:02): If it doesn't sell more cars or it doesn't improve product quality, which will eventually sell more cars, don't do it. You don't have time for this shit. Ford consciously quit working on EDRs for, say, several years until they saw the regulation coming and said, okay, this is going to be become a regulatory thing, so I have to work on it to be able to sell cars. They did go beyond what they had to do, to their credit. That was a few, say, people with good vision in the room that did that. But I think you saw that it's a lot of work. (00:42:44): I always say jobs are easy for the people that don't have to do it, so if you say like, oh, why don't they just release all this data to Bosch? I can tell you when I was in that interface position that it was incredibly time-consuming. While I guarantee the days are better now, I still remember a dark day back in 2003 where I wrote out the specification for one of our modules on a cocktail napkin and handed it to Jeff Weger, the programmer, and said, here's how to read this module family. When Bosch took over and they found some of the cocktail napkins in the records, they said [inaudible 00:43:21]. We need a formal specification. Lou (00:43:26): Yeah. That's not the German way of doing things, you mean? Rick (00:43:29): Apparently not. But what you see is that different companies would make different decisions. At least my observation was that Ford erred more on the side of, well, if we don't have to do it, if it's not regulated and we don't have to do it, why should we be spending time on this? And then there's two different philosophies. One says data is good. First of all, data is always good for safety investigations. The more data you have, the more you can figure out what happened. Now then you get to litigation. Now, General Motors has figured out over time that data is also, say, well, it's good for safety, it's great for litigation. When you get sued, people tend to make up stuff and say, your car did this, that, and the other thing, and if you don't have data to defend yourself, who's to say that what that allegation that's being made isn't true? (00:44:26): Now, GM was kind enough to share that nugget with us at Ford, although at Ford, we had that vice president of privacy that at least initially really put a damper on our thinking of saying, we only record things that we have to have to have to have to know that the system worked right. But when it came time to adding extra data, other than that meeting I was in back in 2006 that resulted into 2009, I haven't seen a tremendous leap forward from my friends at Ford. I've seen a few small steps, but I haven't seen that big leap forward the way that GM has, so clearly GM is in with both feet all the way. Ford is saying, well, okay, we can give you a little bit more. Toyota gave us a big lump in 2013, and we're seeing, say, signs that they have pedestrian modules. They'll give us some pedestrian data. We can get extra data from Toyotas, just not from the Bosch system. (00:45:32): I think you have some people that are afraid that if the product liability bad guys, as the auto auto industry sees them, get too much data, that they'll be able to say, oh, you should have made your system differently. When I was in the safety office but working with assisting and defending product liability lawsuits, all that the plaintiffs had to do is to come up with a reasonable alternative design. Say, oh, you designed it like this. If you would've designed it like that, my guy would still be alive today. Well, of course there's a difference for this accident and for, let's say, 80% of the accidents. The way this manufacturer did it was the better way, but for 20% of the accidents, that other guy's system might've performed better in that one. If you give people too much information, then it makes it easier for them to be able to say, oh, well, okay, yeah, for this accident, you should have had that system. (00:46:32): How come you didn't have that system over there? I think that's a fear that some of the automakers have. Lou (00:46:36): Going back to GM and how they're pushing things, you mentioned it earlier, that forward-facing camera. What do they call it? The front camera module. Rick (00:46:46): Yes. Lou (00:46:47): What are we getting out of there? How often does it record? What do the images look like? Have you seen that data? Rick (00:46:52): Yes. Basically, say, while it's taking thousands of pictures and using that to discriminate the crash, when it comes to the memory, we're storing three photographs. Basically one from four seconds before the critical event of interest, one at the moment of the critical event of interest, and one four seconds after that critical event of interest. I would say that the picture quality is pretty good. I believe the ones I've seen, I think they're even color. Now I'm questioning, but compared to Toyota, Toyota takes like 20 grainy photographs and GM takes three good ones. Lou (00:47:34): Yeah, exactly. Pick your poison. Rick (00:47:36): Yeah. GM, it's been good. In the case of, for example, pedestrian impacts, you can see, I think there was one where the pedestrian was approaching the roadway and then turned at an intersection. They looked like they were going to go right across, but then they turned. The system said warning, warning, Will Robinson, pedestrian coming across road. It took a picture of them going sideways like this. Well, then they turned and there wasn't really a threat, but it took a picture that now showed the pedestrian on the side of the road going along the side. No threat. And then it took one four seconds later with no pedestrian in the photograph and a car a quarter mile out ahead that said, yeah, danger, quarter mile. A car way out ahead. The last photograph was irrelevant, but it did an excellent job. (00:48:30): It was only three photographs, but it told the story. Why did the system activate and then why was there not a crash event that was avoided? I've got other ones where a car broke down and people got out of the car and they're milling around the car. I was actually speaking with Toyota now where we get the 20 grainy photographs and you can see, yeah, there's a car stopped in the road. The light is green and everybody else is going, but they're stopped in the middle of the lane of traffic and people are out milling around and up comes my car from behind. It gets to the point where it says, "Car! People!" (00:49:13): It did avoid the collision and the driver did steer at the end, so you can see the driver, say, because it didn't have automatic steering, you could see the driver steer because the focal point of the camera changed. Even the grainy photographs are doing a great job. The Tesla photographs that are recorded by their video systems are incredibly high quality. You can see the pixels of the stoplight, say, the detail of the texture of the stoplight surface. They're in color and you can see red light, green light, yellow light. I keep saying they're in color. I'm actually wondering if it's my imagination were they in color, Rick. Lou (00:49:57): I think Tesla is in color. Yeah, Tesla is in color. That IPTM, when I saw Don Floyd speak, to my recollection, the photos came in upside down in black and white and you had to... Rick (00:50:06): They were definitely upside down. You had to invert them top to bottom, not rotate them. Should say invert, not rotate. Otherwise, if you rotated, the car appeared to be on the wrong side of the road. Lou (00:50:22): Yeah, exactly. Figure out what country the car was in before you confirm that your rotation methodology was appropriate. Rick (00:50:32): Yes. Lou (00:50:32): But yeah, I think they were black and white, but I don't really remember, either. Tesla is definitely color and the Tesla vehicles and any vehicle with the ADAS, I always say ADAS systems, but I guess the system is already baked in. But with ADAS, they're obviously using the cameras to perform their functions. But from what I know right now, and I don't know what Rick Ruth knows, that's for sure. They're not recording any of that stuff unless being asked to by the operator. Is that your understanding as well? We're never going to get video from a camera mounted on the B pillar of some ADAS-equipped vehicle, or not yet anyway? Rick (00:51:10): First of all, just a disclaimer that I'm not tracking the cameras on a day-to-day basis. I still view the limits of my job as passenger car and light trucks event data recorders. I try to save a little bit of an awareness, but I'm leaving some of that to Alan Moore. It's time for the next generation. You can count on me to remember all the 1994 to 2020 stuff, but the 2020 and beyond, I think I'm handing off to Alan. Lou (00:51:41): Yeah. I spoke with him yesterday. He's obviously a wizard and knows a bunch of that stuff. It does indeed take a village. Rick (00:51:49): Yeah. But I think what you're seeing is that just for sheer memory size and processor time is that you can't write down every picture. They can process every picture for the function of the module, but they can't write it down. If they do choose to write it down, there's a limitation of how many pixels are you going to have in that picture? There's certainly a trade-off, as you say, GM, three clear ones versus Toyota, 20 grainy ones. Which is more important? I think that for we in the reconstruction business, it really comes down to the big things I think are, first of all, what color is the stoplight? I always joke with my classes about, say, yeah, did you see the data in the EDR for what color the stoplight was? Yeah, did you see that one? It's like, oh, you didn't? (00:52:42): Oh, it's not there. When we have somebody that decides to make a left turn, so now it comes down to visibility. What could the driver see at the time they made their decision to do the turn? What could they see? We always try to do time, distance, and we try to do a clock where we say, could the car coming through the intersection, the one that's usually speeding like a bat out of hell, when the driver made the decision, what should they have been able to see? Could they even see the car out in front? Well, a picture is worth a thousand words? If you've got a picture and there's no car visible or they're obscured by other cars when the car begins to make its left, you'd say, oh, okay, they couldn't see enough. Then you come down to when could they become aware of it, and then if so, what did they do? (00:53:40): I always say, when you're in middle of a left turn and you have an oncoming vehicle, you say, okay, I can stop dead in the middle of the road and get killed here, or I can floor it and try to get out of the way, which will work 2% of the time because you can't accelerate a car that fast. Or you can do the deer in headlights and just go "aahhhh" and do nothing, which is what 98% of the people do. And then they get whacked. But boy, if we had that picture with the view ahead, that picture is worth more than a thousand words. (00:54:19): That's really the big hole in my personal opinion. When people present cases to me, I always say, well, what's the critical issue here? Is it a speeding case? Is it speed that caused a loss of control that caused a loss of life? Or is it visibility that a driver couldn't see? A lot of the cases come down to what color is the stoplight? We keep looking for that data element for what color the stoplight is, and it's not there. I've come up with some ways to try to... (00:54:53): ... it's not there, I've come up with some ways to try to infer it. We found that we have certain markers of people that when the light has changed, now they push down a little harder on the gas pedal. Lou (00:55:02): Yep. 100% throttle is generally an indicator that they saw a yellow light at some point. Rick (00:55:07): Indeed, indeed. So, but we really want what color's the stoplight? I see the camera as the ... until we get into interconnected vehicles, and I think they're still 10, 20 years out yet where we have a controller for the stoplights that's beaming out the light pattern, and have the cars picked up on picking up that signal and writing it down in an EDR, that's the 20 years from now EDR. (00:55:34): Until we have that, the camera is the best short-term fix for what color was the stoplight. Also importantly, what was the visibility of the driver of other vehicles that may have influenced their decision whether or not it was safe to turn or not. I'm sure when we do, that you as a motorcycle reconstructionist, that you'll probably see a lot of teeny tiny little motorcycles in the distance coming towards the cars, and then getting bigger and bigger and bigger and bigger. Lou (00:56:07): Yep, "never saw it. He must have been going really fast, 'cause I never saw him." That's what I hear every time. So the Subaru has the EyeSight system, big, apparently extremely effective cameras. That's not coming out in the CDR report from what I understand. Is there a way to get data from those cameras? Rick (00:56:25): Well, some stories from law enforcement have been shared where they have asked Subaru to come out, and if you are so fortunate to live in Orange County where Subaru is not far down the street from you, I'm under the impression that they have cheerfully sent an engineer out to look at some of those pictures. (00:56:46): Now, if you happen to be in the middle of Missouri where it's a couple of plane flights for a Subaru engineer to get there, I've heard that they asked, but we're told, "Sorry, we don't do that." So you can imagine, I mean, if you're a Subaru, you're a small company, you can't have your say specialized engineers just doing nothing but flying all over the country and doing it. (00:57:12): Now, when I was at Ford, that was actually one of my great realizations why I worked so hard to get data released to the CDR system was that we were getting say to the point where hundreds of requests a year just to read the data out. I said, "Well, this is a really poor use of my time. We're using a manager here with 30 years experience just to push the technician buttons and read the data and send it out. I'd really rather let the police or other reconstructionists read it for themselves and write a set of data limitations that'll handle 90% of the cases." (00:57:52): Now then if you have a question on the remaining 5% or 10%, ask away, I might have enough time to handle those. But if I have to go out and get the data ... so, I'm imagining Subaru's going through the same internal process saying, "What? They want me to send an engineer to Missouri and then next week to Alabama and next week to Wisconsin?" It's like, when's he going to get his work done? So, what we all hope is that we'll ask Subaru enough times that they'll say, "This is getting to be really bothersome," and they'll say, "Maybe I should just let some third party read this stuff out, and then they can use the data and they won't have to ask us to help them every time." So when I first left Ford, I strongly encouraged every officer that I taught, I said, "Every case you have, call them and ask them to read it. It might take 1,000 of you before they get the message that it's more efficient if they release it to a third party system like Bosch." Lou (00:59:01): I love it. I hope Kawasaki does that, because last time I went to Kawasaki with a module in my hand in Lake Forest, California, it was clear they were starting to tear their hair out, and were just like, "Okay, we get this so often, this is getting ridiculous." It's like, "All right, well, just give us a tool. We'll use the Denso tool. I mean, I know that's what you already use, and we can do that, just give us the magic decoder ring. I don't want to write bits or anything like that, but I want to read them." (00:59:27): I don't know if you keep up with Jeremy Daly at all, but cybersecurity's a big thing of his, and obviously it seems to be where a lot of people are concerned now that we have these semi-autonomous and autonomous vehicles. If somebody hacks into your Tesla, they could pretty much drive it. So, has that affected the way that we work at all and how we interrogate systems in the way that they're designed? Rick (00:59:48): Well, the main way that it's affecting us is that automakers are having to put in gateway modules. So, the gateway module is like a traffic cop, that when a command comes in and says give me something or do something, it says, do you have authority to make that request? So, it basically checks to make sure it's a legitimate request, and if so, it lets it through the gateway. (01:00:13): Now, how does that affect us? In some cases it affects the vehicle's system, but usually the manufacturer will tell Bosch how to satisfy the security guard that you're a legitimate questioner. Now, where we run into the trouble is when we go to back power. So when we go to back power, we're still using the Bosch system, but the electrical system is dead, and so the gateway module is not powered. (01:00:42): So, when the command comes in from the Bosch system and says, let me through the gate, the gateway is basically asleep at the switch, so we can't wake the guard up to let us through. So what it means is that in addition to back-powering the airbag module fuse, we have to figure out is there a gateway module? If so, what fuses it on and back power the gateway module fuse, or just give up and go find the module and take the module out or plug directly into the module. But it has made back-powering more difficult and it has changed the names of the fuses we have to look for. (01:01:22): So an example in Ford, Ford makes a lot of police cars. So when we were at the Illinois conference, half the police cars in the lot were Ford Explorers, so we started popping hoods up and trying to read them and seeing what we'd find out, and the fuse had evolved from airbag fuse to extended power module fuse. That meant that they kept the airbag module alive for a few seconds, so they didn't have the GM key turnoff problem where the key would turn itself off and then the airbag wouldn't work in a crash. (01:01:58): So the first step was just keep the airbag fuse on, but through now a switch. So they didn't call it the airbag fuse anymore, it was the extended power control module. Now then they put a gateway in, so then we had to find the smart data link connector fuse and back power the smart data link connector fuse. But then there were two of those, so then we had to start guessing which one and do each one sequentially. We figured out that one of them powered the DLC. Good, we needed that one anyway, and then the other one powered the security gateway module that we go. (01:02:42): So anyway, it took it from just look at the fuse box cover and look for airbag, to we need to pull the freaking diagrams for the vehicle. A lot of those black boxes, I should say in the wiring diagram, they don't show you the guts of the module. So you'll just see a wire going in and a wire coming out, and it doesn't tell you what that wire did in the middle. But in some cases that was the gateway module and the magic was it decided whether or not to let you through. Lou (01:03:17): Yeah, as if back-powering wasn't stressful enough already. But I do appreciate not having my car hacked and driven off of a bridge, so I guess I'll take the gateway module. Not that my car is sophisticated enough to be hacked at this point, my 2018 Toyota Tundra. But speaking of the sophisticated cars, and this might be a question better for Alan, but do we know Polestar, Rivian, Lucid, are they going to play ball on any level? Are we going to get data from those guys? Rick (01:03:47): A great question that I don't know the answer to yet. So I was down at IIHS and pulling modules from cars for research purposes, and they had several Rivians down there and I said, "Well, do I harvest this module or not? It'll take me about an hour to get it out from this smashed up vehicle. Well, I'll come back if we ever get to read them." (01:04:12): So what it meant is that I wouldn't doubt that Rivian is thinking about it, because Rivian, while EDRs are not required in the United States, they can get away with selling them here. Should they decide to enter worldwide markets, 2024 in China, you've got to have an EDR at a second higher level, and if you're going to sell in Europe, they're going to honor the United Nations' proposal that's going to require some type of an EDR. So if you want to sell in worldwide markets, you're going to have to have one. (01:04:50): So, the real question for WREX is people have to think globally now, not just locally. So Rivian is fine for today in the United States, but is Rivian intending to go global? I don't really know yet. The Rivians that I saw were monstrous SUVs, so I don't know if they have a real big market in Southeast Asia, but I could say, boy, have they got the United States market pinned down here. Lou (01:05:21): Yeah, they definitely do. That's me and my family. I have three kids, so we're looking for the biggest SUV we can get. So yeah, I talked with Mark Crouch last week. He's a reconstructionist who specializes in some video stuff and he's in the London area. Man, we were talking about the frustrations of those vehicles being equipped with EDRs, but he's not allowed to interrogate them. It sounds like that's changing. (01:05:47): I'm in the states, I'm always so focused on the states, I forget that people across the pond don't necessarily have it as well as we do, and so many of the vehicles don't have black boxes that they can get data from. So, it sounds like things are changing via the UN that you were speaking about, and then China is, you were telling me, mandating modules even when there's no airbags, which I don't know who has cars without airbags anymore, apparently China. But yeah, could you talk a little bit about that? That's interesting. Rick (01:06:21): Well, and a lot of this information comes through Don Floyd, because I was particularly curious to know his experience, because he is their global EDR chairman and has had to be on the front lines of satisfying it. He said that China entered into the EDR world with basically a flawed premise. (01:06:41): Their premise was we're going to do this once and for all, we're going to make one size fits all EDR. We're going to tell everybody exactly what to write down, what memory address to write it in, how many bits and bites it was going to take up, what the resolution was. It was going to be we're going to make everybody do it exactly the same way so we can make one tool. The one tool will just say, dump it and it'll tell you everything in it. (01:07:16): So by making a one size fits all never changes, they didn't allow for any future expansion or addition of any other future data elements. They did it in a way that basically that they didn't want to pay Bosch. I think they were going to buy 100,000 of these things and give one to every intersection traffic officer they had, and the officer was allegedly going to just plug into every car after a crash and say, "You're at fault. You're not and you are." (01:07:47): Now, you and I know from the work that we do here that it takes a lot longer than that most times, that it's not always immediately apparent. Even if you have the EDR report in front of you onsite, that it's like, oh, well, so somebody turned left in front of somebody that was going a little bit over the speed limit, but when they turned left, what could they see? Should they have allowed more time for their left turn or is it all the fault of the person coming through the other way? Who has the right of way? Not so easy. But China just assumed that, oh, it's going to be immediately obvious to 100,000 officers that have not had any training or experience. Lou (01:08:33): That's what we were all afraid of in the early 2000s. We're like, "Oh, they got EDR data, they don't need recons anymore." But like you said, that was clearly not the case. That's not how it works. Rick (01:08:41): Yeah, and because China was so restrictive about how they did it saying it must be done exactly this way, they basically boxed themselves out of using a United Nations global proposal where it could be part of a worldwide solution. So, the China one is going to ... say from what the current take on the market is, is that they will have their own EDR. General Motors, they'll have the General Motors EDR in it too and the China EDR. They're going to have to make a special just for China EDR, and it's kind of wasted money in the sense that the GM doesn't even know if they can trust the China EDR, because who is going to make the tool? GM will not have a contractual relationship with the toolmaker, for example, potentially. (01:09:41): Whereas if Bosch screws up, GM calls Bosch in and says, "Hey, I'm counting on you to meet my regulatory requirements. You screwed up. I'm a big part of your business, it's like, fix it." Whereas in China, if something goes wrong, what does General Motors say? It's like, "Oh, so the polarity's wrong on your GM car? I'm terribly sorry." It's like, who made that tool? Lou (01:10:09): Yeah. Yeah, that's tough. Rick (01:10:10): So I mean, technically China can say, "Well, we told you it had all be the same," but there's going to be some missteps there. Lou (01:10:17): The big question is ... you got most of it, I think you handled China perfectly. My next thought is what's going on in England? Are they going to finally get some additional support and be able to open up that market and download a bunch more cars? Rick (01:10:35): Well, the answer is eventually. So, I actually just had a good conversation in part of preparing for this with a police officer in Rotterdam, the Netherlands, who's ... one of my partners in Europe, the EUDARTS Group does training. They took off from the training that I started and they actually deliver it in every imaginable European language you can think of over there. (01:11:02): But I said like, "Hey, you just got the regulation, and before this you had Toyotas, you had Volkswagens 2018 and later, you had Volvos, you had a few of the Fiats from the Fiat Chrysler group, but your regulation said everybody has to have an EDR if it's a newly designed car." He said, "Yeah, guess what? Nobody made a newly designed car, so we got no EDRs in the first wave of the European regulation." (01:11:32): He said, "Now that's going to change, so another year and a half down the road, we're going to get the next phase of the regulation kicking in where old designs have to be fitted with EDRs and then we should get some more of them." But he said, "Right now," he said, "we're still not getting a lot of them." (01:11:53): So for example, a Ford in Europe, if you put the European VIN in, it says, I don't support European Fords. Say, well, what if I tell you you're an American car? He said, "It'll go and it'll get some bits out, but it might not have a US software ID, so it might hiccup on the software ID and say, well, I'm not supposed to read that one." He said, "Some of them, I'll see them come back and we'll get a few bits. So I know they're talking, so I know that they know how to talk to each other, but I only see this... I know there's data in there, but they're not giving it to us." (01:12:35): So, apparently some of the Europeans have the on off switch. Say electronically it's not a physical on off switch, we're not allowed to have those and don't want those. But they at the factory, they have the ability to program EDR on or program EDR off, so many of the European ones are programmed EDR off. Mercedes, I believe, he was citing was a similar situation. They could communicate, they suspect there's data in there, but they can't get it out with the Bosch system yet. Now BMW, they had a little bit more luck with. Lou (01:13:15): Yeah, and it sounds like, if I'm reading between the lines, it's five to 10 years or so they should start to see a lot more compatibility. Rick (01:13:23): Yeah, and we've had our regulation for 10 years and now we live in a fabulous world where the vast majority of our vehicles have at least some kind of an EDR that we can read, many with the Bosch and the Kia tool. We can say we're up to not just 99% of the new ones, but probably getting pretty close to 75%, 80% of the cars on the road. It's primarily cars before 2012 that weren't General Motors, weren't Ford and Chrysler goes back to '06. But we've got the big four pretty much covered, so we say what's not covered in the United States? It's not the big four and before 2013. Lou (01:14:12): We got a nice buffet to choose from. Like I said, it's rare. I mean, geez, you get a crash now where neither vehicle has EDR on it and then you kind of feel like you're flying blind. It's like a middle-aged, dark-aged reconstruction. I feel bad for our European brethren who are still dealing with that, especially on challenging cases where just a quick download would tell them so much. Rick (01:14:36): I just say, they've made their bed and now they have to sleep in it. They're very privacy-conscious and they didn't want to say invade the privacy of the driver, so they haven't. Now, it's debate whether ... in the United States we talk about what is the public interest and what is the private interest, so we've struck a balance that we think is reasonable for the United States. (01:15:02): In Europe, they struck the balance a little differently than we struck it, but now they've changed their mind. So for many years the balance was on privacy for them, and now the balance is swinging the other way in terms of data that's in the public interest. Lou (01:15:17): I got to get a bumper sticker that says pro data, something like that. Rick (01:15:22): One thing that's interesting is the way the Europeans have dealt with the privacy issue is that they've said that they want to make sure that whatever tool is used does not download the VIN, because the VIN can be traced via databases back to the driver. They want the record to be independent. Say somebody drove that car, but we have no idea who and can't find out who, we can only tell you there was a crash. (01:15:51): They want to not have GPS data to identify the location and they don't want to have time to say exactly when it occurred, because all those things could be used to go look up and say, "Well, who was driving that car that crashed on March the 11th at 3:42 in the afternoon at this intersection?" That you could use that information to get back to the driver, so they wanted to decouple, de-identify it. (01:16:17): Now but as they try to do the balance, they say, "But wait a minute, we also want to be able to study cars by group and say, do people die more in small cars than large cars?" Of course they do. Say do people die more in Corvettes than they do in Prius's? Of course they do, because the way the driver behaves. (01:16:39): But they're struggling with the, well, how do I write down what kind of a car it is for research purposes, but I don't have the VIN? So, you say like, "Well, wait a minute. Now, if you have the ..." So, I'm just going to make up a scenario. Say like, "Well, if your Bosch tool was networked to the internet and you had internet out in the junkyard, you could enter the VIN and it could go look up the type and download it and put that in the EDR record, but then not write down the VIN. Sound complicated?" "Yeah." (01:17:11): You say, "And what you don't have internet out in the junkyard? It's like, well, do you write the VIN down in the CDR record? But then you come back and you go to the database and change it to what is acceptable to put in the record and then you erase the VIN. You say, "Well, wait a minute, how do we know that they erased it?" They don't have to, they could just keep the original record. (01:17:35): So, the Europeans are struggling with how do I this? How do I balance this privacy and the rest? So, the only thing that's pretty universal is they don't want the VIN in the record. We saw that same thing with NHTSA. When NHTSA's researchers go out, they actually had a rule where they couldn't have the VIN in the record. Bosch actually made a special version of CDR just for NHTSA that doesn't record the VIN. Lou (01:18:03): Yeah, and that makes sense, because it's a publicly available database with all the photographs and everything. The VIN being a private issue is anybody can walk up to my Tundra and read the VIN off. Right now, there might be somebody standing over my windshield looking at my VIN for all I know, and what do I care? A, and then B, it seems somewhat hypocritical that they have all these speed cameras on the highways that are capturing vehicles speeding and whoever's in the driver's seat, yet you're not going to let us get the VIN for research purposes to try to reduce how many people are dying in fatal crashes every year when it's such a big issue? That just seems very shortsighted and silly. But what do I know? I'm just a reconstructionist who specializes in motorcycle crashes. I wanted to switch gears a little bit, go through, this is my speed round, Rick. It's not as fast as probably your average guy's speed round, because you and I seem to like to chat, but obviously one of the big points of my presentation coming up is the future, so I wanted to pick your brain a little bit and see if we could use your crystal ball. (01:19:08): I think somebody who started in 1973 and is still going at it 50 years later, is that 50 years? Yeah, probably is in a good position to say, "Well, this is how I've seen it evolve and this is where I see it might go." You might be able to shut down some of these questions and say that's never going to happen. (01:19:24): One of those questions is, do you ever think we'll get video as part of a download, videos of a crash? We're getting photos right now, but it sounds like there are some big technological challenges associated with processing and storing video as part of an EDR report say. Rick (01:19:40): Yeah, I don't see it in the next five years, just because of memory and size limitations, but heck, electronics are getting better every year. They're not just getting better, cheaper and more importantly in smaller spaces, because actually the physical room to put these things into airbag modules. (01:20:04): They ran out of room some time ago and they have to rely on technology improvements to scrunch the box up a little bit to make room for anything else, in particular capacitors to keep it running, to take things out of temporary memories and put them into EEPROM or flash memories. After the crash is over, you got to keep the module running for a while and that data transfer takes time. (01:20:28): So, can I see it in the future? I can see it, but at the moment it's so far out that it's kind of hazy and on the horizon. So these are pure guesses on my part, but I'd say ... and I know people that say something can't be done are constantly being interrupted by people saying, "Well, this guy's already doing it, or this country's already doing it." Lou (01:20:50): Yeah. Rick (01:20:51): But I don't see it happening for at least five years and maybe 10 years until we can develop the capability to be able to store that, and also to make the case that it's in the public interest to have true video. (01:21:10): I think the short-term solution of a modest number of photographs is still a pretty decent solution, that Toyota every six tenths of a second in my mind is better than the GM-40+4, because I think that to me it's a really big deal in the left turn situation to say, when did the other car come into view? With a picture every six tenths of a second, you've got a shot at that. Lou (01:21:46): Yeah, and you got to placate the privacy attorneys as well, so the odds of getting say video of the cabin seems highly unlikely. Maybe they'll even balk at video of the outside world at the time, because you might catch a pedestrian who's robbing a bank on the corner. That's not cool, you can't do that, you're violating the criminal's privacy. What about post-crash data? The big rigs, we're getting 10 seconds of post-crash data or whatever it is. Actually, I'm going to talk to the DELTA |v| guys later this week, but you think there's any chance we'll get post-crash data like past collision pulse? Rick (01:22:12): The short-term issue is power loss and having to keep the module running with capacitors. All those capacitors you see on the boards today keep the module alive for 0.15 seconds, and if you want to keep running for five seconds, that's actually 30 times more capacitor. So then another question is, where do you put that capacitor? It doesn't fit in there. So, I don't see it happening a lot. Now, I've seen several EDRs not covered by CDR that tried to do that. So an example, the 2008 Ford F-250 was set up to go from minus five to plus five. Now they assumed that the vehicle would've power for the plus five, and if the vehicle didn't have power, it just wouldn't populate. The CANplus wouldn't keep running and you just wouldn't get any data coming in, so it'd fill with FFs or zeros in there. So I think it's thought of as a nice-to-have, it's not thought of as a got-to-have. Primarily the emphasis on the public interest is how did the first collision happen and everything after the first collision is collateral damage. Say like, okay, so we rolled 16 times. It's like, do we really do care? I mean, there might be a few liability attorneys that care and say, "You should have- ... a few liability attorneys that care, and say, "Your inflatable curtain should have stayed up for another five seconds so my guy could go 125 and roll his car." Lou (01:24:01): How inconsiderate. Yeah, exactly. Rick (01:24:04): But I don't see it as a high priority, and where I really notice it is in the newer Toyotas that have the 17 EDRs that have the ability to store four front events, four side, two rollover, and they've got six sets of pre-crash data. So in those rollover crashes, I've got 10 events in the memory. And I actually tell people, "When you're analyzing these, don't start at the first one in the memory. That's the end of the rollover, you don't care about that. Go find the first one. Do the speed round through the 10 events, find the big delta-v, the biggest one in the thing." Say, "That's your first event." Say, "Lock in on that one, figure out why that one happened, and then go see, 'Is there anything else in here that we actually need to worry about?'" But I tell people, if you have events 10, nine, eight, seven, six, five, four, three, two, one, it's like, "Well, we don't know if..." Say, if one and two weren't related, I say, "If you find out the first one was three, go analyze three." If you start at 10, by the time you get to three, you're starting a notch off. Lou (01:25:13): Yeah. That's what I was going to say, you've got to be well-rested with a cup of coffee for that. Rick (01:25:17): Yeah, you've lost focus, 'cause you've spent so much time on what is irrelevant, and frankly, once a car starts rolling, everything's just going bangy, bang, bang, bang, bang, bang. There's lots of noise, but not much critical data to help you analyze the crash. But as you start at the beginning of the wreck, and work your way forward, say, first get the cause of the first event, and then if important, work your way forward from there. But don't start at the most recent event, not anymore. That's the old bad way to do it. Lou (01:25:53): That's interesting. And that brings up, actually, the next question that I was thinking about, which I reached out to a couple friends who I trust, and have been in the business for a very long time, and I said, "Hey, I've got a chance to pick Rick Ruth's brain tomorrow, what do you want to talk about?" And the perpetual fear, and I think it's been in existence for 25 years is, well, more advanced EDRs are going to eradicate the need for us pesky recons, but as you're talking about a file like that, the more advanced the data it seems the more advanced the recon has to be, not the opposite. But what's your take on video, and ADAS, and advanced EDR systems? Is that ever going to put us on the sideline, where we've got to go start flipping burgers at In-N-Out? Rick (01:26:35): No. No, it's going to come out just the way it's come out for the last, say 15, 20 years, that it's more data that's going to take you more time to analyze, it's going to increase your billable hours, and you're going to have a moment - just as when we had momentum analysis and then EDR, we had to say... Well, now we have to say, "Do they agree?" And if they didn't agree, spend time on fitting them together to figure out what was wrong with the EDR? Or what was wrong with the momentum? Did we get the drag factor wrong? Did the tires leave the roadway? Were we in a yaw? You had more things you had to think about, it was more billable hours. And the other thing we're going to see is that the data's going to become overwhelming to the layman. So even some of the more recent Chrysler reports, and some of the more recent Toyotas, have auxiliary sensors in them, data from the side satellite sensors, and the front auxiliary sensors. Well, we don't use that data in reconstruction, but it's more pages in the report. I'm now dealing with 70, 80, 90, 100-page EDR reports. And out of all that, it still comes down to probably three pages that make or break my analysis of that file, but I've got to sort through 97 pages just to come up with what's important for my reconstruction. So I think the skill that the reconstructionist will need is to more quickly sort through the chaff, discard the chaff, and get the wheat from it. Now, in the case of the videos, once you find the video, and you find the million dollar shot, the one that we all yearn for in our inspections, the million dollar photograph that shows this defect, or shows that bolt that was loose, or that part. It's the same way in the video you're going to be looking for the million dollar shot in the video, and if you find it, great, more power to you, that's going to help prove the case. But I don't see it ever putting us out of business, I see it's going to require more sophisticated, say reconstructionists, that are able to comprehend what they are looking at. Lou (01:28:53): Yeah. And that's exactly what my experience has been so far. If I get a case now that has EDR data, it has surveillance video, then that case is going to take me a really long time to make sure that all of my traditional calculations are in agreement with the video analysis. That has to be performed independently, that has to align and agree with the EDR analysis, which is performed independently. And it's, quite frankly, just gotten to the point now where I can take way fewer cases than I used to. The analyses are just more in depth, more complicated, so I just have to throttle back how many cases I subject myself to. And that's been a learning process, because I'm still probably taking too many cases. Rick (01:29:32): Well, what I think you're seeing is that putting all the data together is the challenge for the modern reconstructionist. What will the reconstructionist look like five to 10 years from now? They'll have 10 times as much data, but the job will still be to fit it together. So I mean, just as a quick example, I get a lot of calls where people say, "Help me understand this EDR file." Let's say, okay, here, say, I've got two different levels of service, one is we put it up on Zoom, and I walk you through it, and say, "Well, here's what I see, and here's what I think is important. But now it's up to you to figure out, does that match your post-crash travel? Does that match your momentum? Does that match your witness statements? Or you want me to write a report." If I'm going to write a report and put my name on it, we start at 10 billable hours, and we work our way up, depending on how much stuff there is. Because if I'm responsible for the final conclusion, then I need to know all that other stuff, not just the EDR. So I offer, say, the one or two-hour consult, which, for most reconstructionists, is what they need, because they don't have the specialty analysis that I do. But somebody's got to put it all together, and that's not an easy job. And the more data that we have, the more things we have to crosscheck to see that they all do fit together. So I think that is the modern reconstructionist, it's almost like you're the big picture manager, and you might to end up doing consults with the drone guy, and the original... The guy who shot the scene with the conventional thing, and the witness interviewer that the attorney's hired, and you're taking all this stuff and putting... The momentum guy, the crash simulation guy, the Terry Day guy, the Berla guy. Who can do all those things well at the same time? Very few of us that I know of. Now that's why I also see, you've seen the engineering firms getting a, call it a bigger hunk of the market share, that they can have an office with 10 people in it, where one does the ED crash simulation, or another one does the Terry Day simulation, and another one analyzes the drone footage, another one analyzes the berla data. I'd actually like to know how many people actually can do all that themselves, and if so, how much do they charge? Lou (01:33:23): Yeah, exactly. Their overall invoice is going to be huge, and if they have the talent to be able to accomplish all that, their bill rate is probably pretty high. I mean, you've just hit perfectly on a theme that has come up throughout all these podcasts. I think my first podcast was with Jeff Muttart, and that was his big takeaway, is, "Hey, man, at the beginning of my career, I could do the whole recon, it's just not feasible anymore, there's too many parts that come into play." Obviously he's not doing recon at all anymore, he's just doing human factors, which speaks to that same concept itself. It's going to be a multidisciplinary effort, or you're going to have to take one case a year, and try to attack every single angle of it yourself. I love it. This came from Jarrod Carter, one of my good friends, and a great reconstructionist out of Washington, and he was saying, "Is there any chance that this ever gets standardized, as far as what's recorded, how it's pulled from the vehicle, and how it is reported? Is there any chance?" I know you're involved with the ISO, and ASTM, and SAE... Okay, tiny little chance. So you're saying there's a chance, but it's just very small? Rick (01:33:34): Well, so first of all, I mean, we do have the United Nations, they have no power or authority, but they have made an effort, they've come up with the UN 160, call it... Say, EDR proposal, to say, "Here's what we recommend." So if you're Jamaica for example... I teach in Jamaica, and I have a good personal relationship with them, and they say, "How do we get all this data here in Jamaica?" And they say, "Well, you need a regulation." The problem is you're Jamaica, you make a Jamaican regulation all by yourself, the big companies are going to say, "Who do you think you are telling us what to do just for your country?" So what is the power? The United Nations has the power of the pulpit to say, "Listen, if anybody needs an EDR regulation, here's a model regulation we recommend that you adopt." Now, what we've always run into is that you always run into somebody who says, "Well, that's fine, but I need something special. That's not going to work for me, I need my own." And that's China. So China said, "Wait a minute, this stupid regulation doesn't capture rickshaw crashes." It's like, "You can't have an EDR that doesn't capture rickshaw crashes." It's like, "This is a five mile an hour delta-V is stupid." So you're going to have people that do that. And then you've got the privacy interests and the public interest people fighting it out. So you've got the Europeans that are saying, "No VIN." And you're got the United States people that say, "Well, the VIN is the whole way we set the tool up. To key off the VIN to know how to read the data in the car, so how do I read the data in the car if I don't know what the VIN is, or something to tell me what I'm reading?" So besides The U.S., China, Europe, there's... Minorly different regulations in Korea, Japan. In Korea, if you don't read the EDR out when somebody asks you to, corporate executives can be sent to jail. It's interesting nuance to the regulation. Lou (01:36:00): That sounds intense. Rick (01:36:01): So you'll still see, call it vested interests, in different countries. And I think what you really see is that regulations, unfortunately, are largely made by bureaucrats. And I don't mean to pick on the... I live in Washington, D. C. now, so I'm beginning to gradually understand the mentality, but bureaucrats think they're intended to regulate something, and they have, in their mind, what's important, and they don't always ask the people that are involved in the industry. Now, when I was at Ford, I was offered up to go to NHTSA and explain EDRs to them, and everything, but at the time that I did that, my impression was, all they cared about was that they had to cost justify, and say that this thing was free. And so that I got 5,000 questions on the cost of each little bite and bit in the thing, 'cause they wanted to say it didn't cost anything. And I was trying to say, "You need this data to understand crashes." And they said, "Don't bother me with that stuff. I'm a regulator, I need to write a regulation, and the regulation has to be cost-effective, so let's grill you on cost for another hour here." And sounds like the GM tried to help China to go down the right path of, "Here's the data that you need." But China said, "No, I've got my mind made up. I want this one-size-fits-all EDR, and I want it to not have to pay Bosch. I want to be able to have a Chinese company do something, and I want it to be a no-brainer. You just take this data element from here, and interpret it that way, and that data element from there, and interpret it that way." So if you say, "Are we ever going to get the whole globe together?" I'd like to use and fall back on my Ford experience, back in 1981, Ford went to make a, I forget if it was a Fiesta, or an Escort, but a small car. And we swore we were going to have it the same between The U.S. and Europe, 'cause that was the way it was going to save money, if we could have a common platform between the two. So we started down that road, everything was common. Along the way, every little provincial engineer in their own little chimney would say, "Well, the way those guys did... We need just a little different. Over here, we don't have that. We don't make any of those nuts over there, we need this nut over here." When we got all done, say there's something like 15,000 parts in a car, there was one bolt that was left common. Lou (01:38:47): Okay, so it's a really small chance. Rick (01:38:51): The only way it works is if you have a power figure at the top. So at Ford, when Alan Mulally came in, he came in, and he said, "What kind of problems have you guys had?" And we said, "Ah, we should really all have global platforms, but we can't seem to make it happen." He said, "Well, why not?" And we said, "Well, everybody just decides they need to have their own little special thing." And just another trivial example, but in India, if you have an Escort-sized car, well, the guy who owns the car, if he can afford an Escort in India, he's the boss, he sits in the backseat. You want a limo-sized backseat for him. The driver is hired for 50 shekels, nobody cares about the driver. But in this country, we want to have a big front seat, Texas front seat for the driver, and there's nobody in the back, so we don't care. So try to make one vehicle common between India and the United States, different market, different demands, very difficult to do. So anyway, Mulally came in at the top, and he is a very humble man, he basically just said, "Common platforms, make it so." And said, "Well, we've tried that before, but it didn't work." He said, "Make it so." He said, "Anybody wants an exception, tell them they have to come see me, and I will tell them, 'Make it so, make it common.'" To this day as they say, he says, "I didn't come up with any of these ideas, I just listened to all you people, and did what you told me." But he made it happen, because he had absolute authority within Ford Motor Company. But now we say, "What do we do about global EDRs where we have, say 25, 50, 100 manufacturers worldwide, who has authority over them?" The answer is no one. So the United Nations has the power of the pulpit to suggest, and if enough people adopt the United Nations' suggestion, it will tend to take off and become wider. But we still see so many regional provincial interests that say, "I need something different." So that's why the chances are this big. Lou (01:41:13): We might be left with, instead of one bolt that is common, we'll be left with one bit that is common, and everything else will be different. Rick (01:41:21): We'll have speed data and delta-v. Lou (01:41:26): Yeah. So, I mean, like you were saying before with the Wild Wild West, that was your term that you used before we started recording, and then I might have stolen it during this conversation, but- Rick (01:41:39): You did. Lou (01:41:42): I'm admitting it now. If anybody's still listening, they'll get the truth at this point. So ultimately, it's up to the reconstructionist to have their finger on the pulse, to know what the right tool is for the job, to know what data they might get from X, Y, or Z, and that's a big undertaking. So ultimately, we're all going to have learn how to stay educated, how to keep our finger on the pulse, and how to adapt. Do you have any recommendations for us? The best way to stay in tune with what's going on? Rick (01:42:12): Well, I mean, frankly, it's go to your regional and national conferences where people appraise you of what's new, what do you need to be thinking about this year, what's new on the table. So I've just used the example in EDRs, it's the Andy Rich and Bob Scurlock paper on adjusting for the EDR location. For 20 years that I was in the business, we just said, "Ah, EDR, it's sort of on the center of gravity, almost, kind of, sort of, pretty close." And for the funnel and front angle crashes, it was close enough, but there was that occasional crash that was on the front bumper or the back bumper that was a T-bone fashion, where you say, "Whoa, it makes a big difference there." And when I started in EDR, I used to say, "There's no crash that I can't explain." I was a little pompous. You know how young people are, those 50 year olds, those youngsters, they'll get a little pompous, and they think they can do anything. And then you stay in the business long enough, you eventually get one where you get it, and you say, "Right, these numbers don't make sense." Why don't they make sense? 'Cause I'm using my old conventional hat, and I haven't yet learned the new tool of adjust for the module location. So you say, for that one in 50 cases where the module location makes a difference, so we say, what is it that the reconstructionist needs? He needs to have the awareness of all these idiosyncrasies and tools just enough... He doesn't have to be able to solve everyone personally, but he's got to have the awareness to say, "Something funny's going on in this one, and I know it, and I may not have practice in this one, but I've got to throw this one to the specialist to say, 'Help me out here, 'cause there there's something weird going on.'" And that's the skill that I think is going to be the hardest skill. In our information age of too much information, it's going to be the people that have either the gut feel, that says, "Something's wrong here." Or they've been to enough conferences, they've heard about the different things, to say, "Oh yeah, I remember hearing at one of the conferences, 'You might have to adjust for the module location.' And they said it was only a few crashes." And he said, "Oh yeah, yeah, this is one of those. Oh, okay. Yeah, okay, we're going to need to either go research that, or go study up on it, or do it." But I think the challenge, is the reconstructionist has to be the big picture vision of, "What's important in this one? What do I need to focus on in this one? Are there any special circumstances that I need to consider for this one, that maybe in 90% of them are just plain vanilla crashes, don't even need to think about it?" Who can be that person that decides, "This one's plain vanilla, and this one is tooty fruity gumdrop. Lou (01:45:28): Yeah, I like those tooty fruity gumdrops. Yeah, exactly, I love how you brought all of that up and put it together, and it's something that... In my career, I started really... I probably went to my first crash conference there in Vegas with Scott Baker in 2003 or 2002. And I remember just showing up to that conference, and being like, "There is so much information I have to know." And you kind of have to steepen it for a decade before you start picking up all these little things that you need to be aware of. That's me. I'm probably just a slow learner. But the bottom line is, you do indeed have to go to all these conferences, you have to speak with other reconstructionists, you have to take classes. Without doing that, you're not going to have... That word, I love that you used that word, you're not going to have the awareness to know when you're in trouble, and when something funky has reared its head. So I will just recommend taking your course, take one from Rusty Haight, and you told me this during your class, take one from Rusty Haight, take one from Brad Muir, take one from Kent Butz. Find whoever, go to conferences, listen to Adam Hyde from Northwestern talk about it. Anything you can do. And right now, you're teaching a class, from what I understand, at SAE and IPTM. Rick (01:46:49): Yes. At IPTM, we have a level one and a level two. Level one for, call it geared towards your law enforcement officer just getting into it, and level two is for your state cop that's been in it for five or 10 years, but has run across those weird ones where they need the specialty stuff, but they can't go hire somebody, they have to do it themselves. Lou (01:47:14): Yeah, exactly. Rick (01:47:15): So level two is a- Lou (01:47:18): Well, yeah, that's one of the big challenges, is the poor law enforcement officers, they might not have the opportunity to accumulate this multidisciplinary team like us private guys might have the luxury of doing. But sorry, I cut you off. Rick (01:47:29): No, it's okay, I was just going to say that the level two is typically triangular velocity vector method, it's offset collisions, it's ground forces, it's now adjusting for the airbag module location. We used to also throw in there all the new EDRs that... And we found that we couldn't cover that all in a week, so we decided to keep it to the, call it the skills, what are skills that have to be developed to be used, as opposed to just awareness of what the latest data is. So that course has evolved over time. And the SAE course started off as the, called it the three-day engineer version of the IPTM, five-day law enforcement. The five-day law enforcement, we would make the officers do the math in class, because if they didn't, they wouldn't do it when they got home. And in SAE, I'd say, "Well, would you like to do the math in class, or would you like to..." We'd just show you one. And they say, "Well, Rick, unless you're paying us 300 bucks an hour to do math, we're not doing math for you, Rick." It's like, "We do math for our clients. You show us once, we go do the math at home, we charge them for it." So my SAE classes, when I said, "Who would do this drill sheet?" And say, "That's okay, just show us one." Lou (01:48:48): Yeah. Rick (01:48:49): So we would cram it into three. So initially, it was crammed five into three. Well, now that there's a level two IPTM, I've got 10 days of stuff, and I'm trying to cram that into three days for SAE. And I keep asking SAE for a fourth day, and they say, "Well, we can't call it a class if it's four days, it's a seminar. It's not a class anymore, it's a seminar. Can I have a seminar?" Lou (01:49:17): Yeah, exactly, they make it a seminar. I took your class. I was looking at my notes in preparation for this conversation, that was 2021, so it was mid-COVID, so I had to take the remote version of it, but you did a tremendous job compiling all that information and putting it into a format that's digestible and concise. And then I also just really appreciate the fact that you gave a bunch of materials that you didn't necessarily go over, but now I have on my computer, and whenever I get one of these Tutti- Frutti cases, which is what I'm going to call them from now on for the rest of my career, I can pull up some information from the PowerPoint presentations, and I have that. Like I said at the beginning, you're a pillar of this community, you really help us interpret all this information. And I appreciate you donating two hours to speak with me today and to the rest of the community. Much gratitude. Thanks, Rick. Rick (01:50:15): Very good. Lou (01:50:18): Hey everyone, one more thing before you get back to business, and that is my weekly bite-sized email, To the Point. Would you like to get an email from me every Friday discussing a single tool, paper, method, or update in the community? Past topics have covered Toyota's vehicle control history, including a coverage chart, ADAS, that's advanced driver assistance systems, Tesla vehicle data reports, free video analysis tools, and handheld scanners. If that sounds enjoyable and useful, head to lightpointdata.com/tothepoint to get the very next one.