This interview was conducted in 2020.
Background
Two connections. The Amherst Fire Department is relatively new to this, however our State Fire Marshal through the Department of Fire Services has had drones for the past three to four years now. I’m a member of their Special Ops teams. So we go to responses around the state with various different assets. And the drone program was the newest asset that they started about three, four years ago. They have acquired about half a dozen drones that are kept around the state in different locations and if a community needs one for an incident or pre planned event, then one of us will go and get it and fly it. And then on top of that, we’ve actually applied for and are waiting to hear on a grant. So hopefully we can buy one for our Department here in Amherst soon.
We’ve been using drones mostly in the fire service, certainly for the Fire Marshal. We’ve been using what I’m going to call the lower end commercial, upper end consumer models. Your average fire service does not have a lot of money. There are some big departments around the country that can afford 50k drones, but what we’re dealing with here are things like the DJI Inspire, DJI Phantoms, the DJI Maverick Enterprise. The Maverick Enterprise is actually the platform that I would be buying here for Amherst because again, the cost is something within our reach but it offers simultaneous regular imagery, thermal imagery, and the loudspeaker is actually something we see as useful for when we’re doing searches, which we do run into up here with lost hikers in the woods, about half dozen per year in the Amherst area. State wise SAR stats are much higher.
Also, I attended FDIC 2019 and encountered a UAV platform that offers some advantages over their current platform. The exhibitor was FotoKite, see FotoKite (link to website) & FotoKite (LinkedIn About). The advantage of the drone being tethered is that you don’t actually have to have a pilot license to use it. So we could have any of our firefighters deploy it. And secondly, because it’s tethered, it has essentially an unlimited flight time because it’s taking its battery power either off a vehicle that it’s permanently mounted on or a vehicle that you’ve plugged it into. So for a lot of our uses, like a search and rescue out in the woods, anything like that, you need a true drone. But for a lot of our uses where we’re trying to get footage of an active fire scene or something like that, we don’t really need to be flying around all over the place. We just need an eye in the sky, so the kite concept is that you pull it out of the box either onto a tray on the vehicle or a Pelican case and put it on the ground. You deploy it, it goes straight up. I think it’s either a hundred or 150 feet tethered. So there it’s filming, you can move the camera around, but the physical drone doesn’t move cause it’s just straight up on a tether. I would say for 75% of what we do for us in the fire service is filming straight from above, that’s all we would need. You would still need a true drone for doing other things like search and rescue wildfire or some pre-planned events stuff. But it was very intriguing for that because of the unlimited flight time and again, because you don’t have to have somebody with a license to fly it. But my understanding is they’re quite expensive compared to other drones.
What types of missions do you fly with drones?
The main ones I can think of both pre preplanned event flights and during the event, that I’ve been involved in. One event took place a year ago. We have a lot of large County Fairs, like state fairs, things like that. One event took place near the Berkshires where they wanted to get a feel for traffic flow during a typical busy day, how the parking lots swell and go down again, because they’re working on trying to improve both access in and out, especially if it was an emergency and they had to move a lot of people in and out quickly or get emergency vehicles in or out.
We also fly missions to show stakeholders what kind of traffic patterns develop, what kind of parking issues develop because it’s so spread out. You can only see one small piece of at any one time. So the team actually had a remain up there for a day and basically fly the drone every hour and get snapshots and video of what it looked like. Throughout the course of the day you could see lots and the traffic swell and then go down again.
Then actually going back a step, they went and filmed the area before the fair started for that season just to get the lay of the land without any fair setup. And then during the event we would fly every hour just to get updated footage of what it looked like throughout the course of the day.
Then we turned around and gave all that footage to the stakeholders that run the fair and the local public safety to analyze data for the following years to show people what the flow looks like.
For that type of event and most of what I’ve been involved with, it’s manual flying. Some of the units that we have for the State Fire Marshall does have the Pix4D software. We’ve just started to really train on that more. Because with some fixed points on the ground, they can use that to get a lot of footage and then feed it into a computer and get 3D models after the fact.
Most of what we use it for, we’re looking for live instantaneous, either aerial shots or aerial video, to see what’s going on in an incident as it is unfolding. The State Police use Pix4D for accident reconstruction and now we’re starting to use it for the fire service, after a major fire to image the building that burned down. So that’s the only sort of automated piece we’re doing. Most of what we’ve done up until this point has been manual flying and just trying to keep the same track each time that we do it.
Which missions to you fly the most?
- Examining lay of the land prior to events such as State Fairs
- Flights during the event to monitor flow of traffic
- Recently, conducting inspection flights for fire services after major fires
- Responses to state of emergencies (caused by extreme weather) environmental emergencies
How much do you fly at night?
Yes. a couple of the drones that we have do have thermal imagery. We have done that and that’s been used quite a bit actually during fire incidents or search and rescue. Those are the two main ones where that comes in. Nighttime we’ve applied for and are waiting for a waiver from the FAA so that we can do night flying and do overcrowd flying without having to get permission on a case by case basis.
It’s rare that we’re trying to get permission for on an emergency scene basis because the timeliness isn’t there. We need to either do it or not do it. We don’t have time. That’s not quick enough to get approval from the FAA on a case by case basis for true emergencies. But for pre-planned events, they’re obviously very sensitive about any airport proximity, but in terms of just in general with no airports involved, either nighttime flying or crowd blind, they have given permission for preplanned things that somebody wants to have done. We just want the blanket approval now, especially for an emergency, it gives us that extra flexibility to do it at night or closer to crowds.
How will having >1 hour of flight time improve your workflow?
Not having a long battery life is probably one of the single business biggest shortcomings on the drones we operate at the moment. One of the more recent times I flew was for a staged hazmat drill, at one of the state’s reservoirs and they were simulating a spill into the reservoir and tracking where the spill goes and how they could boom it and how effective the booms were. So the outside company that was overseeing these operations wanted the drone to provide an aerial view from above, see the spill as it migrated and how the booming was affecting it. But we could only stay up there for 20 minutes and then I had to get the drone back again, reset it, send it out again. So we had to keep doing that versus having a drone out there for a longer period of time.
Doubling what we have now would be phenomenal. Like most things, the longer, the better. You know, a lot of the time you don’t need that long of a flight. But for some things like that, a longer flight is certainly better. I think an hour would sort of be a golden standard to try to reach, if we were able to keep these drones up for an hour versus 20, 30 minutes, that would be a huge, huge improvement.
How important is size to your use of drones – where do you have to store and deploy the drone that drives how big you need the drone to be?
On the state side of it, they have ISU incident support units deployed around the state, and those are large RV type vehicles that are mobile command posts with radio assets, conference rooms and other things. So what they’ve done is they’ve actually deployed drones with each one of those incidents, support units around the state. So they’re basically kept in cases inside those vehicles that way. There’s one in each part of the state. So they’re strategically placed around the state and they can either be deployed with that vehicle or just the drone itself depending on the need of the incident. So in terms of size, it’s not like it has to be small, but portable is good, but it’s not like it has to be teeny tiny fit in the backpack sort of thing, because there is room in those vehicles to store such assets.
How important is high resolution performance overall and high resolution low light performance?
We certainly want it to be good enough for live stream so we can figure out what’s going on and are able to make out different things. We’re not trying to do high res images, you know, photos that we’re going to use. Something like hazardous materials incidents, they’re wanting to get close enough to let’s say a tanker truck to be able to read from a distance, for example the placards on the vehicle. Usually those are reasonably good size, but things like placards on a vehicle license plate, you want to be able to pull that out. Most of your drones these days are capable of that with the resolution that they have.
Is an HDMI output or ethernet output from the ground controller important to you?
I’d say it’s very important. And in terms of things where we are looking to see some improvement, whatever the best way that can be, you know, come up with to get that live signal to another larger monitor or fed into the internet. That’s one of the areas we were exploring more now because one of our drones, the Inspire, has the ability for two controllers, you could have a controller for the drone and another one just to the video. And that one you could have in a command vehicle and fed into a large monitor or whatever. So people could see it. All the newer, smaller drones that we have, the Phantom, the Maverick, others, they only have a single controller. So you’ve basically got whatever the small 8-9 inch screen is in front of you, but that’s it.
How important is sound signature to your use of drones?
For our needs, it’s not, it’s not huge in terms of the operation. We’re no in any way trying to do any kind of covert surveillance, so having a quiet drone is not directly a need, but indirectly, especially when we’re doing something like when we are flying around crowds like that fair example would be great. Then quieter obviously is more important just because then people aren’t taking notice of it. We’re not trying to be covert, but at the same time, people get nervous when they hear the buzz of a drone overhead. Um, so in those regards, quieter is certainly better.
How important is US made to your purchase decision?
There’s two parts to that. From a security standpoint, it’s definitely a concern. Again, we’re not conducting any top secret covert operations where we’re really worried about somebody hacking in and seeing what we’re seeing. But at the same time, it has come up, whether news organizations could try to hack into something and get your live stream – that’s always a concern. And what would be worse is if someone is able to hack in and take control of it and cause damage, injury, etc.. So that concern is certainly there.
The other piece of the Made in the USA that we do occasionally run into in the fire service industry is a lot of cities are big on made in the USA from a union standpoint. You want things that are Made in the USA – you want the money staying in the USA – they don’t want to be buying products from China. I’ve heard that come up also.
Is there value to your workflow in a truck launched drone platform where you never have to leave your truck and can fly over people and inspect areas of interest and items of interest?
What it comes down to is anything that’s going to make the deployment easier is potentially beneficial. The need to stay physically in the truck and fly it, at least for us, is not necessarily a big benefit. But the ability to have very quick deployment, if that means pulling up, opening a compartment or a case and be able to pretty much fly right out of the box without a lot of assembly on any given day, that’s probably the most important. Any of our incidents, unless it’s preplanned, we are always going to be short-staffed. There are never enough people to go around, which is actually one of the obstacles if you will, of a department the size of ours from having a drone at fires.
For something like the State Fire Marshal, that program works well because whatever community’s having the fire or the incident, they’re not the one flying the drone. They’re calling the State Fire Marshall and saying, we need a drone here and now. And then two other people, me and somebody else as an example, show up and fly the drone for them. We have nothing to do with the rest of the fire or the incident. The downside is it could take an hour or more to get us there.
The upside of us having our own drone is it’s right here in our town – ready to go. The downside is the first 15-30 minutes of a fire, I won’t have enough firefighters to fight the fire. I can’t dedicate two people to fly in a drone. It’s kind of that give and take of the best way to do it. So anything that’s going to ultimately speed up, getting the drone up in the air is going to be a benefit to us. And again, that’s why the tethered concept is also attractive to us because it wouldn’t have to be a trained person to fly the drone. I can have pretty much any of my firefighters do it.
Do you need zoom/ high resolution from a distance to perform your tasks?
Yes, zoom, certainly because even given the costs of the smaller drones that we’re talking about, they’re still expensive. And whether you’re talking about fire, heat, fumesor you’re talking about hazmat and chemical clouds. We want to keep the drones out of harm’s way, as much as possible. Obviously a drone is more expendable than a person, which makes it very useful for something like a hazmat, getting close and getting a field of view. But obviously the further we can keep the drone away for potential harm and not contaminate or destroy it, that’s going to be beneficial. So the better imagery we can get from a distance, the more benefit that’s going to be.
How much do you expect to pay for a highly capable EO/IR small drone made in the USA?
4-5k. However, software and livestream capabilities are intriguing because then the cost can be divided between departments such as Information Technology and streams could benefit several other county and state departments and roles within such as Conservation and Public Works departments, which have expressed interest recently in such capabilities.
What are specific data workflow/ data management challenges you face? How could your workflow be easier?
We already talked about one of the demands that we have now is for live streaming, the auditability to get that signal to another platform easily. Another one of our obstacles is we’ll go do an event for somebody, it could be preplanned, it could be a fire incident for another community. We get all these images now on photos and videos. Now they’re on the SIM card on the drone. So now we take that SIM card and typically in most cases we actually don’t have a way to do anything with that at most incidents. If one of our incidents support units is there, we can then download that to a PC, which takes some time. And from there you can re-upload it onto a thumb drive to give to a Fire Chief, Police Chief and Incident Commander or whoever it may be. But with the amount of data we’re talking about, it takes a long time and you’ve got to have a laptop or a PC there to do it. So that’s definitely one of our challenges is we’ve got to get the images from that SIM card somehow to another portable device that we can then give to whoever the who’s going to take custody of that footage.
Do you have any data security challenges? If so, what are they and how could the drone manufacturer solve them better?
For us, no, because we’re not handling criminal evidence like the way the police would. Fire investigations can go that route. And then some footage that we take could ultimately end up in some kind of a trial because you know, criminal charges. So they do have to be careful with the chain of custody of it and the security, that data, which is why we are concerned about what happens with that SIM card when it comes out of the camera, where it gets copied to how it gets transferred from one person to another. It’s definitely another area that needs improvement because right now it’s an old logistical issue. You’re out in some community, you’ve got a SIM card. Now what do you do with that data? How do you get it to the people that need it?