Episode 189: Precision in Practice: First Responder Tracking & Training with Andrew Power

Episode notes

The best way to train first responders:

Hello Smart Firefighting Community!

Welcome to another episode of covering real world innovations via interviews with fire service and technology industry experts that empower YOU to develop your very own Smart Firefighting strategy!

Thanks for joining our 16 part FRST Challenge Mini Series recorded at the competition! Tracking and situational awareness has long been a problem for first responders. That's why the FRST Challenge competition is designed to pinpoint 3D tracking technologies that provide one-meter accuracy for first responders in indoor environments, crucial for enhancing their safety during rescue operations.

The 6 finalist teams competed in localization and ruggedization tests at Muscatatuck Urban Training Center from October 23-27, 2023. Winning teams gain access to funding, business support, and live field testing opportunities to evolve their solutions for real-world application.

In this episode:

Explore Ravenswood 's breakthroughs in Time-Space-Position-Information (TSPI) for ultra-realistic training scenarios

Discover how a multi-layer TDMA network can support tens of thousands of live entities for comprehensive simulations

Learn about the synergy of open-architecture systems and modular software in enhancing first responder capabilities

Hear from Andrew Power - the Production Manager at Ravenswood Solutions .

Andrew manages Ravenswood's Production and Training as a Service programs, including proposals, project management, personnel management, regulatory compliance, delivery and sustainment of products.

So click play now to hear what he has to share, and stay tuned for the rest of our mini series!

Head to www.smartfirefighting.com to discover how SFF accelerates innovation for emergency responders, to find out when our next event is, or review our curated resources!

Facebook | Instagram | Twitter | LinkedIn

Chapters

1. 0:00 Intro
2. 1:12 Meet Andrew Power of Ravenswood Solutions
3. 1:24 Tension and acceptance at phase five
4. 2:05 What changed since phase four
5. 4:06 Designing for fire, law and EMS
6. 5:12 Surprises at phase five
7. 5:57 The ruggedization session
8. 7:00 The road to commercialization
9. 8:37 Ravenswood and SRI International
10. 10:10 Lessons from defense work
11. 11:42 Designing for the user
12. 13:57 Building trust through training
13. 14:45 Vans and the mushroom shirt
14. 15:53 Final thoughts
15. 16:57 Outro

Questions answered in this episode

What is the FRST Challenge and what were the phase five requirements?

The FRST Challenge focuses on producing marketable prototypes that demonstrate indoor localization tracking of first responders within one meter accuracy without any pre-deployed infrastructure. Phase five increased the challenges by including multiple responders, longer durations, and longer distances.

Hear it at 0:00

How did Ravenswood Solutions improve their tracker from phase four to phase five?

Andrew Power says they implemented more inertial tracking that activates when vision quality drops below a certain point, such as in smoke environments. They also added an infrared LED instead of a visible headlamp so the camera can light up rooms for vision tracking without betraying position in security situations.

Hear it at 2:19

What was included in the ruggedization testing at phase five?

Andrew Power describes 10 minutes at 320 Fahrenheit in an air fryer, then straight into three feet of water for five minutes, followed by a four foot drop and an eight foot drop. The tracker had to run before, during and after and continue to work.

Hear it at 6:02

What is Ravenswood Solutions' background and core business?

Ravenswood is a subsidiary of SRI International that spun out around 2015 because their National Guard training work became profitable. Andrew Power says their core focus is instrumented after-action review, so people can train like they fight and then immediately review that with high fidelity ground truth data and learn from it right away.

Hear it at 8:52

Why does Andrew Power think training is a good starting place for first responder tracking technology?

Andrew Power believes training gives users a lower-risk way to build trust in the technology. A trainee can go through a scenario and five minutes later review an instrumented after-action review with ground truth, learning in days what might take weeks, and that builds trust before extending it to real-life emergencies.

Hear it at 12:01

Transcript

Machine transcript of the published audio. Names may be misspelled and some sections are missing.

Ep 189 transcript

Precision in Practice: First Responder Tracking & Training with Andrew Power

Speakers: Kevin Sofen, Andrew Power

Source: Whisper medium.en machine transcription of the published Podbean audio (no original transcript found in Drive), lightly cleaned. Speakers were separated by voice clustering and then checked and assigned by hand at every speaker change. Gaps where the model returned no usable text are marked [inaudible m:ss].

[0:00] Kevin Sofen: Hello, everyone, and welcome to a new Smart Firefighting mini-series created during the phase five of the FRST Challenge, hosted at the Muscatatuck Urban Training Center. The FRST Challenge focus is to produce marketable prototypes that demonstrate indoor localization tracking of first responders within one meter accuracy without any pre-deployed infrastructure. Many of you heard stories from phase four, and here in phase five, the challenges were increased by including multiple responders, longer durations, and longer distances. In these conversations, you will hear from many of the entrepreneurs, first responders, and team members involved that address these challenges and opportunities of X, Y, and Z axis tracking from multiple perspectives around scalability, usability, affordability, and more. This challenge is administered by the Indiana University Crisis Technologies Innovation Lab and funded by the NIST Public Safety Communication Research PSCR Division. Also, consider putting the NIST 5x5 Conference on your calendar, located in Chicago, June 25th to 27th, 2024. Enjoy this deep dive into first responder tracking technology, and please reach out with your thoughts. In the depths of our mini-series here at FRST phase five in Muscatatuck, Indiana, alongside Andrew Power with Ravenswood Solutions. Andrew, how are you doing today?

[1:22] Andrew Power: Good to be back. Happy to be back at Muzzy.

[1:24] Kevin Sofen: it's really great to be here and I will say just kind of seeing the magic in the air of the kind of the slight level of like tension and angst amongst the competitors while also having all these different stakeholders from DHS, NIST, XPRIZE, just sort of like that magic in the air you see when all these people get together.

[1:41] Andrew Power: Yeah I think there's this balance between anxiety and then just acceptance you know maybe at this point as we get towards the end the trials go how they go. It's cool our common operating picture gives us a pretty good idea how we're performing live so we don't have to kind of cross our fingers as long. We worked really hard to get here from phase four and so we're we're happy to be here and get it done and look forward.

[2:05] Kevin Sofen: Well I know that this a lot of what I see from Ravenswood is different from phase four to phase five. What were some of the lessons learned from phase four and from what I'm seeing this week and kind of what I'm holding my hand right here what's what's different in phase five?

[2:19] Andrew Power: Yeah so there was kind of two categories on the to-do list you know one was stuff that we wish we could have done at phase four and the other one was stuff that we learned at phase four you know edge cases where performance wasn't good because of something about the environment and the trials so example there would be you know smoke environment we kind of saw that coming we didn't have a ton of opportunity before phase four to test like that but we went into that fire trainer in phase four you know and the vision tracking just you know just dropped and so said what can we do and in this case you know we implemented more inertial the tracker itself figures out the quality of the vision tracking it's getting if that drops below a certain point and starts to use more inertial input and then in the same vein the inertial sensor does a little bit of training while you do have good tracking so it's counting and measuring steps while you do have good tracking so that it can interpret better in case you lose vision. Another one, you know, dark room, whether from smoke or anything else, it's challenging on the vision. So I don't know if you remember at phase four, but we saw some of that the first day. And so we went back that evening and we glued a headlamp from Walmart on the front of it. And then that was kind of okay to do at phase four. They said they didn't want us to do stuff like that at phase five. So we actually put an infrared LED with a window inside because one of our police partners at the trial said I would rather not have a headlamp sticking off my chest while I'm searching a building and I said that sounds fair so the camera is sensitive enough to infrared that you know effectively lights up the room for the vision tracking without having to betray your position in case you're in a you know a security situation where you don't want to do that.

[4:06] Kevin Sofen: And here at phase I mean there wasn't the governor house and I think we there was that there was this ruggedization test but from the additional people and even I noticed at the hospital yesterday there was law enforcement and EMT and then here you've seen firefighters. How have you approached and sort of what's your what's some of the design philosophy and around looking at not only fire but also the law in EMS?

[4:33] Andrew Power: Yeah probably number one is just the most universal simple kind of mounting system that we can have you know decent security knock it off not get them caught up not have them stand there for 10 or 15 minutes while you're trying to sew it onto their vest or whatever and then you know just not burden them any more than we have to we don't want stuff to get in their way to weigh them down as we you know look forward keep getting that form factor down and down and down, make it look like something they recognize so it's easy for that user community to kind of accept it as a piece of instrumentation rather than something scary and different and new.

[5:12] Kevin Sofen: So from the experience and from the Ravenswood team, I mean, any sort of, anything surprise you or kind of catch you by surprise at phase five in your past couple days here?

[5:24] Andrew Power: Most surprising was how dang long this tunnel is. know we said we're going to the prison complex again I said okay no problem you know we pretty much tracked almost all the way through that tunnel last time comms shouldn't be a problem and they went the other way with it and went half a mile down underground so eventually we did lose comms but it tracked pretty far down that tunnel and that's probably the only real surprise wasn't surprised I would say as we performed a little better than we even expected to and the ruggedization testing we made it through the whole oven session tracking the whole time.

[5:57] Kevin Sofen: We haven't talked about the ruggedization session. Could you explain what that was?

[6:02] Andrew Power: Yes. So there were several parts. First we went 10 minutes at 320 Fahrenheit and literally air fryer, you know, off the shelf air fryer. And the key with the ruggedization was not just visibly broken or not, but you know, it had to run before, during and after and continue to work, which I think is, you know, is good for realism. that we went straight into a garbage can full of water. So call it three feet deep of water for five minutes and then a four foot drop at an eight foot drop. So we, that email came out not too long before the trials and we said, wow, that's, that's a little more severe than we expected. So what can we do, you know, in the next couple of weeks? And we didn't do a whole lot, but I was still overall pretty pleased with the way that the units reacted to the testing. Yeah. Certainly that's something we'll consider as we get, you know, closer to a product that would be for sale.

[7:00] Kevin Sofen: So on that note of for sale and I guess the roadmap to commercialization, I mean, it's October 2023 and the phase five FRST Challenge is coming to an end. What's next and how are you approaching continuing to try and take this and get into the hands of first responders?

[7:15] Andrew Power: Yeah, absolutely. So, you know, and I said this to the judges last night, you know, it's end of the challenge, but it's not the end of the journey, the project, the product. Obviously we have stuff in mind about what we want it to look like and how big we want it to be and other performance specs and reorganization specs. The step from here is hope that, you know, our performance in the challenge, things like this podcast, meeting people at this event kind of drums up enough potential business where we can create a theoretical business case for ourselves that says, okay, let's spend this much money making it a really good product based on I can sell this many at this price afterwards. And you know, we're lucky that our kind of core market with DoD supports, you know, high player counts, right high volume. So the way I look at that is if this is going to cost $1,000, you know, to throw out a mega made up number, if I can sell 50 of them, then it's going to look exactly like it looks today. If I can sell 50,000 of them, it's going to look a lot more like an iPhone, you know, or something small and compact and sexier, you know, basically. So it's try to make that case where we can put really intense resources, time and money into making it an awesome product that just works right off the shelf.

[8:37] Kevin Sofen: Yeah, that's great. And then I also, just for maybe the context of people who don't really know Ravenswood Solutions and even kind of the whole company back story, but kind of give me a little background on Ravenswood and then sort of kind of the alignment you have with some of the other partners.

[8:52] Andrew Power: Right. So yeah, so Ravenswood is a subsidiary of SRI International and the reason that, and so SRI is like the, what's called the original Silicon Valley Research Institute. The reason we split out is because the work we were doing with the National Guard grew and grew and grew and became profitable. And that doesn't work with SRI's classification as a not-for-profit. So around 2015, we spun out and our core, which is to say one-legged table at that time was training, National Guard training. And we had a couple other small programs with that was say 90% of our revenue. We've diversified quite a bit since then, but still we're pretty focused around what we sell, we would call instrumented after-action review. So how you get the data can vary, you know, what's the instrumentation, but the idea is we want people to train like they fight, and then immediately be able to come back and review that with high fidelity, you know, with true ground truth, with a good user interface where they can see and understand that data, played back to them with short latency. and then learn from that, you know. So you skip the part where you're trying to remember who's right about what you did or didn't do, look straight at the ground truth and then learn from it right away.

[10:10] Kevin Sofen: So you've got some past DoD framework and knowledge and experience from a customer standpoint and I'm sure a lot of your team, how has kind of that experience of working in the very deep and closely with DoD help translate or how have you used that or leveraged that to continue to apply that for first responder solutions?

[10:30] Andrew Power: Couple ways. I mean, one, even our army tracker, dismount tracker, the name Freddy is an acronym that had first responder in it. So state of South Carolina came to us around 2011 and said, you know, we need situational awareness during disaster response. So that was that first dismount kind of size tracker that we made. So we kind of been tuned in with that. Another one is just, you know, a profound respect for the work that's being done by the user community. So, you know, with the army, I don't look at it as being better at fighting, but less loss of life all the way around by being trained better. And so that obviously connects with the, you know, the first responder community. Like let's save as many lives as possible through use of technology, whether it's augmenting training or live, you know, active duty situational awareness. And then through experience, I would say, with these communities is understand how important and impactful the user experience is. The best tracker in the world is worthless if nobody wants to wear it because it's annoying or it's hard to use. So really focusing on that design for the user.

[11:42] Kevin Sofen: I love that focus. And I want to even go a little bit deeper around the change management aspect. I mean, what have you experienced, or what are your thoughts in regards how they get end users to actually do something different and to actually integrate new technology to ideally make their jobs better but also it is making a difference so that's a challenge.

[12:01] Andrew Power: You know I was talking to Ari this morning too and he's like in my, he was a medic for a long time and he's part of the first team and he's like in my experience it was see something save your life or save somebody's life and you kinda hang on to it. There is a lot of momentum you know and it's I think the paradigm has changed a little bit generationally and younger people are a little bit easier to embrace new technology but really what they have to do is they have to just see it work and see it help them. The help to them has to be tangible. Certainly I'm sure in our careers somebody comes in, a consultant to somebody who said hey we're gonna use this new efficiency tool at work and you have to sit through an hour seminar about how it works and I said that's extra work to me right. You have to see it tangibly help you, I think, to accept it. And so, you know, even if this initial rollout of this type of project from us is in a training kind of niche within firefighting, I think that's kind of a great path to acceptance, right? Because you would see as a firefighting trainee, cadet student, I don't know what they call them, but somebody who's ready to become a firefighter, goes through a training scenario and gets to sit down five minutes later and look at an instrumented after-action review and learn something about what they did that they didn't notice because their nerves were high, their adrenaline was high, they ran up and down the stairs, did whatever they needed to do, but being able to go back and play that and replay it with ground truth and understand, and they said, I learned in a couple days what might have been a couple weeks how to do this right, there's some trust built on that and seeing that the map that they get back is reflective of real life it looks like the way they went through the building build that trust then maybe you can start extending it to saying hey this will help me if I'm in trouble in real life.

[13:57] Kevin Sofen: Yeah I love that perspective and that's the sort of I've done a lot of work within virtual reality training and there you're able to record the scene and be able to play it back and similar to what you just said within training towers the ability to kind of have that Monday morning quarterback perspective in terms of people in playback, why did you do this, why did you go there, and what was going through that decision-making process. I do agree that I think training situations are gonna be a good place to see this technology in the crawl-the-walk phase here, and hopefully further adoption in the next couple years.

[14:28] Andrew Power: Yeah, I mean, give them the opportunity with a lower-risk way to start using it and build some trust in it, and at the same time, still make the product better, right? Because it's gonna be brand new onto the scene. I'd rather have not maybe lives at stake for the first year that this is being rolled out, for example.

[14:45] Kevin Sofen: Yeah, well said, and I would be remiss to not slightly pivot the conversation, but to first compliment your shoes. I grew up wearing Vans.

[14:53] Andrew Power: Thank you, me too.

[14:54] Kevin Sofen: So rep the Vans, but more so the shirt. You're wearing a mushroom shirt. It's the shirt that I would be wearing, but you're wearing it. Why the mushroom shirt?

[15:02] Andrew Power: You know, it's a fun looking shirt. I just figured people would remember me if they had talked to me today. say, did you talk to Ravenswood? I don't remember. Did you talk to the guy with the mushroom shirt? Oh yeah, I did talk to that guy. I remember that guy. Yeah, and you know, and they're a fascinating, interesting kind of life form. As a gardener, they helped me too.

[15:20] Kevin Sofen: In the gardener point, another point you said earlier, but the power of mushrooms, and one thing I think in looking at the network underneath and the whole idea of mycelium, I think that kind of last night from when everyone was collaborating, working that sort of a, maybe a sort of a maybe a little representation of kind of the power of measures and power of networking.

[15:39] Andrew Power: Yeah, interconnectedness and communication, absolutely.

[15:42] Kevin Sofen: And just got to keep it rolling, so I appreciate the context and the mission.

[15:47] Andrew Power: Yeah, I'll retroactively say that's why I wore it, for the interconnectedness.

[15:53] Kevin Sofen: Yeah, I'm all for it. Well, I guess kind of final thing, if I could give you a chance to say a final mic drop of any comment, a question, a quote, an insight, a challenge, you know, anything that you could leave us with here today.

[16:06] Andrew Power: Wow. Oh, that's putting me on the spot. I don't know. We just feel privileged to have been here. It was such a lucky chance that we met Sunny a couple of years ago. It's been awesome to meet and participate in this community. Certainly igniting a passion, you know, to really want to do a great job with this product looking forward. And yeah, just opened my eyes to kind of different ways to incentivize good development between getting your name out there, you know, cash prizes, getting to meet people, going to cool conferences. We had an awesome time at the 5x5 this June. So yeah, I don't know. I don't maybe have profound advice for the next team, but just kind of like a thankfulness to everybody who put this together and for our chance to participate.

[16:53] Kevin Sofen: Well said, Andrew. Forever grateful and look forward to continue to collaborate in the future.

[16:56] Andrew Power: Cheers.

[16:57] Kevin Sofen: Thanks for listening to today's episode. What did you think? What'd you learn? Hopefully something, right? Please reach out on social media with your thoughts on this episode and any other suggestions for future content. We look forward to capturing more important stories across public safety and are here at your service.