# Parachute Recovery Systems for Safe Flight over People

- Event: NestGen ’22
- Published: March 4, 2022
- Length: 34:12
- Watch: https://nestgen.org/library/watch/BdJHEMn9Cu0/

> The transcript of a recorded NestGen session, for reading or for use with an AI tool. It comes from the video's
> captions; automatic captions can mishear names and terms.

## About this session

Failures can be minimized, but not eliminated. Learn how AVSS are helping make drone crashes virtually harmless with parachute recovery.

## Transcript

**[0:13]** failures can be minimized but not eliminated learn how avs's are helping make drone crashes virtually harmless with parachute recovery welcome to session number 15 of NestGen 22 parachute recovery systems for safe flight over people to give you some context before I hand it over to our speaker here today um I often imagine a day when I look outside my window and see several drones flying over me delivering packages monitoring road traffic preventing and fighting fires spraying seats and whatnot in reality there's a lot that needs to go behind the scenes to make that happen as d-rich pointed out in our keynote address it needs to be a wide set of contingencies that we account for while building an autonomous system it goes without saying that fail-safes aren't set in place with the belief that things won't work but rather to avoid situations that aren't under our control to affect us negatively we need to essentially give the system the flexibility to adapt to different situations enabling it to function without human intervention drone parachute recovery system or brs and flight termination system have been

**[1:16]** proven to be some of the most reliable fallback mechanisms designed to reduce the risk and downtime related to flight failure their prs is equipped with internal sensors that continuously monitor flight conditions in real time autonomously deploying a parachute when a failure is detected with backup options such as manual control and alarms the system is designed for high safety and compliance with civil aviation authorities abs is part of a global network of drone technology stakeholders that have welcomed the requirements for third party testing and validation for prss to speak more about this of course we have co-founder and ceo josh ogden he is a graduate of the university of new brunswick a single entrepreneur and a peer-reviewed published author before helping start aerial vehicle safety solutions in 2017 josh worked as a technology and regulatory management consultant with a diverse group of early and mid-stage technology companies at avs's josh is responsible for leading the long-term vision of the company and assisting the engineering team with building products that meet the current and

**[2:19]** future regulatory needs josh is also a member of several drone industry related groups including the several astm f38 committees and the canadian rpa's r d task group thanks so much for being here with us today josh over to you thank you very much for having me in the introduction thank you very much for joining us today hopefully it's not too early not too late for everyone um so again my name is josh I'm the ceo of avss we started this company in 2017 with a simple question of what happens when a drone fails one thing to to recognize over the last couple years safety has become a key component of any flight over people or beyond visualize site and that's where jake stokey says it great the drone parachute revolution is here and why because of safety regulators want an environment where safety is considered and there's redundancy systems built in and today's session so today we'll be just doing a brief overview about avss we're

**[3:21]** going to discuss why we would use a parachute and then we're going to go into key features of a parachute from there we're going to end with going into some practical but nitty gritty insight of the parachute recovery systems in some different markets such as the united states and europe so about abs avss started in 2017 and is now a global distributor of parachute recovery systems we have more than 50 authorized dealers selling our product across the world we are also our product is compliant with the third-party standard astm f-3322 this is a compulsory test for flight over people in many of the countries and this requires crashing a drone 45 times to demonstrate the performance of the system we've done three astms and we've actually got three more scheduled for the spring I think by then we'll have the most astms completed in the world some of our customers who are using our product to enable more are telecom providers construction companies and public safety as well we have a host of very high-end

**[4:25]** dealers such as adorama globe flight rmus and advecture selling our product additionally for the m 300 and the canadian market for example we are the number one choice with 85 percent of the market and we see that similar representations in other markets such as europe recently and excitingly we get to announce a new product that we uh are launching and received a contract this is a guided parachute system which I'll actually be discussing towards the end of the conversation and then so why use a prs a parachute recovery system you know when we speak to drone manufacturers or end users it's a really common theme that you know there's there's been a lack of trust with the parachute manufacturers historically because of their robustness and their performance as well it's an expensive item that adds weight so ultimately why are you going to use something that's going to reduce your flight time the key thing to understand is the regulatory environment as I hinted too at the start civil aviation authorities have recognized that drones are coming especially since since covid however safety is paramount they don't want things

**[5:28]** falling out of the sky so our product is enable an enabling technology to reduce the kinetic energy and help with people flying beyond visual line of sight and operations over people for example in canada if you were to buy the DJI m300 or the DJI m 200 and you wanted to use it for over people you wouldn't be allowed however if you add in abs the modifier as you can see on the screen here you're suddenly enabled to fly over people under the advanced category and the reason for this in the canadian market is a single source of failure cannot cause the drone to come down and the kinetic energy thresholds um and some examples some practical examples so this is a customer of ours called altimax they're a DJI dealer as well as a service provider for offshore and oil and gas as well as utility they were doing a mission across newfoundland where they were counting utility poles and they were going in the backyards of residents and to do that in canada they would needed a parachute recovery system for that flight over people

**[6:29]** additionally in canada we have an example at the 108th gray cup so this is the canadian football league our lesser version of the nfl however during this event um typically you're not allowed to have drones flying over however with a parachute recovery system and in canada what we call an sfoc a special flight operator certificate there the customer mohawk college as well as the hamilton police they were able to put an m300 above the crowd during this public event with nearly 20 000 people another example is vladis aerospace they are a company that is across north america and I believe south america and they were doing some beyond visualize site operations across canada and for them to be enabled and approved to do that again a parachute recovery system for that extra redundancy the extra safety for the regulator another example that we see parachute recovery systems being used is drawn as a first responder in the united states so it's chila vista very well known public safety organization they have drugs on the roof that are operating seven days a week

**[7:31]** 10 hours a day and they are a customer who are using parachutes to receive their co-as their certificate of authorization waivers and then finally in europe what we're seeing is our first light uas certificate approval in hungary this is geo layer their approval allows them to do their own risk assessments and enables them flexibility in the european union to go and fly and control it and again a parachute recovery system being a key component to that aspect um and so now after talking some use cases I'm gonna discuss some key features of a parachute recovery system we find that a lot of end users and even manufacturers aren't fully aware of all the feature sets and don't understand that you know a parachute recovery system isn't just you know a parachute that's going to deploy out there's a lot more features in engineering in it and and to start off a key thing with a parachute recovery system is astm f3322 compliance so there's a lot of misinformation out there on what is the third party

**[8:31]** standard who can do it and how is it validated there's a new astm standard actually being published in the next 60 days however with the standard what we under need to understand is it requires 45 crashes the new standard there'll be an increase 10 or 15 more tests however you're demonstrating that your system is reliable and in this testing we're checking for single motor failure we're checking for full motor failure we're also doing shock load tests to validate that everything stays together and one thing unfortunately and I think this plagues the drone industry is some companies have gone out and tried to find a rubber stamp testing agency someone who can just say they passed the test without a proper accreditation and this is why we use an FAA designated test site in new york called new air and this validates and this is important um for the new standard that's being approved and and launched the next 60 days or the revised standard that is is that third-party test sites are going to be scrutinized more regulators have been burnt already when companies go out and try to circumvent the the spirit of the of the standard and this is where

**[9:34]** the standards are becoming more robust which ultimately for the end user for other drone manufacturers this means you should be validating and questioning who did your testing can I see the report if they're not willing to share the report and share the details I'd be very skeptical and so again I think that's a very paramount feature for parachute recovery systems when we think of flight over people and beyond visual line of sight next key features is a bracket design one thing that we grapple with as a parachute recovery system company is we're not the primary we're you know we're somewhat second-class citizens and because of that we want to make sure that we're out of the way of the key sensor so if you have a top payload for the m300 you know let's not use that real estate let's give that to you however your your bracket needs to be strong enough to handle an opening shock test so during the third party testing you need to actually drop the drone from for three seconds it's going to approach terminal velocity the parachute is going to open there's going to be a lot of opening shock and so if you don't design your bracket appropriately it's not going to work additionally you need to think about the end user the end user

**[10:34]** doesn't want to be sitting there using a bunch of tools to set up and and tear down every time so there's some usability there so so a bracket is a key component where is it easy to use and did it did it withstand the third party testing next is a flight termination system this is a topic that seems to be a lot of confusion in europe um in the other markets it seems more straightforward however with the flight termination system the ultimate goal is to stop the propellers and you want to stop the propellers for two reasons one is to avoid entanglement you don't want to get caught up in the actual parachute additionally you want to make sure that you don't cause laceration so when the drone comes down and if it were to hit somebody you do not want the propeller spinning now to do a flight termination there's a couple options and so on the left hand side of the photo you can see an external hardware fts for the tb 55 batteries for the DJI m200 this is cutting the power between the battery and the drum and then on the right hand side you see a software fts for the m300 that we also provide that is actually

**[11:38]** sending what they call a kill command through the DJI osdk so again with the flight termination system some regulators are are happy with the software fts some require hardware so there is depending on countries that you're in and their understanding of the flight termination system next is the auto trigger so an automatic triggering system isn't just simply you know gyroscope that you know you flip over and the drone fails that's really simplistic and you can result in false positives you want an automatic triggering system that is independent of the drone and that provides you as the end user or manufacturer also insight into why a failure happened on the screen here you see an m300 that has failed and you can see around the the middle of the circle that's the main the majority of the flight envelope however as you exceed the green which we call saw thresholds which we start analyzing because you know maybe your drone tilts a little bit over however it corrects itself well that's not a failure so that's what we call soft thresholds and then a red the red is when it

**[12:41]** completely flips over and as you can see here here's a real life example of a failure um next we're going to discuss the actual practical nitty gritty so one thing with parachute recovery systems yes it can be there to help save your drone it's not guaranteed if you plan your drone on a concrete or on some rough surfaces things are probably going to break so again if someone tells you from a parachute company 100 guaranteed the drone's in perfect shape I would be questioning that but there is some really nitty-gritty details that I'll be discussing for each country and looking at you know what does it actually mean so to start off in the faas the usa there's a lot of confusion on how you get to flight over people the rules were implemented and we're starting to see progression under the part 107 rules but we're not quite there yet because of category two and three um requirements and that first requirements is means of compliance which no no means compliance has been approved yet that involves a parachute recovery system for

**[13:41]** flight over people we've got one submitted our own custom means compliance that'll be we believe it's three to nine months away I'm waiting on the FAA and I believe they were waiting on the new third party standard but a key thing to understand is in the united states it isn't 11 foot pound or 25 foot pound of kinetic energy it's the severity comparison and that's something that I think a lot of people miss you're not looking in like canada and europe where you're doing a kinetic energy formula what's your descent rate what's the mass of the object and typically you're around you know 60 70 joules which is you know around 50 foot pound but in the united states is we actually have to do additional testing which we call atd testing for the comparison injury to a rigid object so what you see here is some screenshots of our atd testing system and what it is is you're taking the drone you're sending it into the atd the crash test dummy you're looking at the data on what injuries you know neck compression for example occurred and then on the right you have a rigid object which has certain specs that you have to meet and you're

**[14:42]** looking at what would the injury comparisons and if the injury of the drone were less than the injuries of the rigid object under either cat 2 or cat 3 you're then compliant and there's a couple more steps to do but ultimately this is the key point for the means of compliance and so if I'm a drone manufacturer not only am I looking at the descent rates I'm also looking at well what's my drone made out of and what's you know plastic versus metal so what am I going to create for injuries and this is something that you know the us is leading on and is more robust than other countries which does add additional costs but it gives you initial insight into your drone um next is for the dnr durability reliability testing so in the united states there's four categories category one two three and four I just spoke about category two and three category ones for smaller drones but then there's category four this is a rather new development in the united states when they originally launched the rules for flight over people this wasn't included but once they received comments during the notice period they did actually introduce a

**[15:45]** category four this is for type certification so the philosophy around type certification is if you have more data on the reliability of the system your drone for example you meet certain flight hours the severity of injury is allowed to be higher because the likelihood is less and this is where category 4 comes in and something that's quite often overlooked by drone manufacturers who are entering into the type cert world is that a parachute recovery system can actually reduce your time to market so as you can see on the two columns on the left hand side your hours for example if you want to fly in a rural area is 375 but if you want to fly with a parachute system that's astm approved and the kinetic energy is less than 128 foot pound or 170 joules you can actually reduce it from 375 to 150 hours and another example is you know in san francisco which is we would say a population density of 20 000 per square mile you're going from 7 200 flight hours to 3 600 which is a considerable saving when you consider what that per hour cost is

**[16:46]** and again this is something that's quite overlooked by a lot of drone manufacturers just because of the process to get into the actual durability and reliability testing next in europe so with europe why does a parachute and flight termination system matter well for pdras the predefined risk assessments under the specific category and one thing that's required for example near airports is you need a flight termination system so that your drone doesn't leave your operational area but you also need a parachute for human factors so one thing that that you know we believe is quite clever and and quite um intelligent by the european union was they recognized that you know some pilots don't want to just have their drone fall out of the sky because it's leaving the area they might they might hesitate they might delay so they included that parachute system to provide you know well at least you might not lose your complete job now that's not guaranteed as I said you know during third-party testing we see you know half the drones come out with no damage and then you know 40 with some damage and you know you're gonna lose a couple drones during these 45 crashes however the regulators recognize that maybe it will help these end

**[17:50]** users be more inclined to make sure their drone doesn't leave that operational environment next is ground ground risk class so this is where I'm excited to talk about our new product that we've been working on so when you're flying your drone and using a sore application there's a ground risk buffer rule and right now the rule is one-to-one from our third-party testing and data collected that actually isn't accurate it's it's more of a one to two which means if you're at 100 meters you should anticipate almost almost a 200 meter drift with you know a parachute that can't be guided and so we've actually got real life examples where during our third party testing for the m200 you know as you can see on the right we're at 120 meters and when we deployed it exceeded it drifted to 197 meters so what that means is if you're flying bb loss over people you should be accounting for this as well we're recognizing that the gyrus rules need to be updated for the one-to-one rule with unguided parachute systems but then comes the question on what happens if you could guide it and that's where our announcement on our recent contract with indigenous

**[18:54]** services can in transport canada for our guided parachutes is an exciting part because we'll be launching later this year a new product where we can actually control where the parachute drifts after deployment so what does that mean that means you can control your ground risk buffer zone and reduce it so in the green is an example of us being able to control it a one to zero point five so if I'm at 100 meters being able to control within 50 meters which is a huge opportunity when we look at the drone delivery market for urban cores for example this is an example in japan of a drone delivery route that 7-11 is doing that they announced in december and when you take a look the red is the true drift area that if you were to use a unguided parachute it deploys this is how far it could drift out to but the the yellow is the rs rule which again we don't agree with because of data supporting that will drift farther and green is the new guided parachute system that we've developed so we believe that this is an example of an evolution in the drone industry that we're going to be seeing um and we've seen yasa

**[19:59]** talk about steerable parachutes for a larger system and we think that this is going to become status quo in the drone industry um thank you very much for attending today uh I know that's a lot of information on regulations and getting into some of the details so happy to answer any questions or if you'd like to send me an email feel free to email me at josh.ogden at abs.co great thank you so much for that yeah that was quite a lot of information yeah so let's just wait for a couple minutes um I guess our audience would like to ask a few questions we have a few uh on our social media channels and on discord but I'll just wait for a little bit longer in case people want to ask questions oh and I'm just going to throw up a video here great yeah that would be great and I can show her a on our guided parachute system

**[21:03]** so here's an example of us doing some drops so again our use case for this technology um initially we'll be doing medical delivery in northern canada this is an area where it's hard to reach there's no available infrastructure for existing drone delivery so they want a system that can do uh critical medical supplies uh without having to invest in the expensive infrastructure so this is giving them an option and showing some landing accuracy of of critical items this also has applications for wildfire for military and other applications interesting and the reason actually for this development was starting in 2018 we started noticing our drones were drifting into trees and power lines and we wanted to reduce that risk and so we started working on the control algorithms that autonomously direct the system straight to a predefined gps point and again I can just replay that video out if there's any questions feel free to put it into the chat happy to

**[22:04]** answer or again you can email me at josh.ogden at avss.co you mentioned that you allow um you don't you don't want to interfere with when when people want to add payloads for example on top of interest degree drones like the m300 now when you're encountering a failure right like how do you prioritize um so typically so when we design the bracket our goal is to land as close as possible to on the feet um and so that's kind of the trade-off sometimes manufacturers will ask well can we land it upside down that sure that's an option however you must be aware that your third-party testing is going to be quite expensive if you're always landing your drone upside down and then I can share a video of our main in-market system for the m300

**[23:05]** so this is the uh a single motor failure actually so you'll watch the drone here it'll start spinning out of control and then the parachute will deploy excellent okay so we're starting to get questions as well perfect okay so uh one of our questions is also around the m300 they've asked how much weight and height do your solutions add to a drone like so the weight is 900 grams you're going to lose about 10 to 15 of your flight time as well as height I assume you mean minimum deployment altitude so that's an interesting topic that our system has worked at 15 meters however the third party testing standard has a requirement for a minimum deployment altitude calculation which is based on your outlier so if you do 45 tests you get your average kinetic energy your

**[24:07]** descent rate however for minimum deployment altitude whether it's hover or full forward you look for your outlier which then set rates your minimum deployment altitude one thing with that we've brought up is that if I'm in a hover state versus a full full forward state the way in which my parachute reacts is different and we believe that there's a revision that will be required so that if you're in your hover state for example delivering a package that your minimum deployment altitude should be rated based on the hover test not on the full floor test so that is kind of just some inside information on some issues with the current standard interesting uh so they have a uh there's a question on how the guided prs works and how it's controlled yeah so the guided prs system is similar to a paraglider where you're pulling the lines of the parachute and it's controlled via gps so you set a gps coordinate the sensor suite inside knows where that target is and aims the uh the parachute the pair parafoil towards that target

**[25:10]** okay and so how would you manage missions in gps constrained environments so we're working on some technology with uh that'll work in not in gps denied environments but that'll come out at a later date okay do you fly rtk gps uh depends on the drone you don't have to uh we have you know on the m210 rtk we have customers flying with the rtk and customers without so it really depends on the end user but we would adapt when we integrate in different drones we make sure that all the full features are available to the best of our ability okay okay you spoke a little bit about uh severity comparison comparing injury to a rigid object and could you tell us a little bit more about how that's different across countries is that defined differently they have different thresholds to that we have someone who missed the earlier part of the session yes

**[26:15]** yes so in most countries it's the kinetic energy formula it's you know you look at the mass of the drone and you look at the descent rate and it creates a certain amount of joules or foot pound that's common in in canada and europe where you're seeing typically you know 70 to 80 joules of acceptance however in the states one thing the FAA commented was that they've observed data where the DJI phantom could create 120 foot pounds of energy but still only create injuries under 11 foot pound so the FAA came up with using a comparison it's a very it's a more complicated system it provides benefit for those drones with a lot of foam I guess around it but ultimately what it's looking at is 11 foot pounds of a metal object is going to create xyz injuries and a drone coming in at the descent with your parachute is going to create xyz injuries if those injuries on the drone under a parachute are less than that of the rigid object they deem it safe so this is something that they

**[27:15]** announce the means of compliance it's coming each company can submit their own we started working on it in 2020 and you know our documentation package is a 200 page documentation and it took a year and a half to build out so it's not something that you can just go do and I've seen some videos online where people are trying to just like drop objects and it's quite a scientific process you know the people that we're working with on this our partners they um test military helmets bicycle helmets and and it's something that you just can't go do on your own like it's a full um system and there's a lot of biomedical engineering behind it interesting any other questions I think you have one other question um what other hardware oriented fail-safe mechanisms have you experimented with how do they rank up against a prs what do you mean hardware orientated uh so I suppose this is mostly a a bracket system that deploys the parachute and

**[28:24]** so yeah so we use a similar fake face yeah we use a high-end uh compressed spring that has about 30 joules of energy and 12 inches in full extension that protects the line as well as it ensures that the parachute goes beyond the tumble zone um we've we've worked with pyrotechnics we think that for the larger drones you know when you're up to 20 25 kg a pyrotechnic solution is probably a preferred option just because of the weight of the parachute and the size of spring to create that energy however for the smaller drones we're a fan of the spring because of the travel ability to travel we have a lot of customers who are getting on planes they don't want to deal with pyrotechnics as well as shipping when we descending to our other dealers so you know when you see a pyrotechnic solution I I would say that's more favored for like you know 20 25 kg maybe beyond visual line of sight kind of drone delivery applications but if you're looking at end users who need to get up and go like the m300 or work tool you don't like a pyrotechnic just adds complexity to your life and your life's complicated enough with all your batteries so we don't think

**[29:24]** you need to add more problems by using a pyro for the smaller drones okay and what about a buoyancy option in in addition to the prs system have you considered that yeah we've looked at airbag type solutions the problem with that is drones are very weight sensitive you don't want to add too much weight to anything and so it would be a nice to have but the frequency in which you use your parachute hopefully never to add that extra additional weight it's something that can be considered and you know if someone really wanted it we could entertain it but it's something that I think at the end of the day it's nice in theory but when you look at the cost and the weight of the system you would be kind of saying well do I really need that it's kind of a it's a nice to have I would say and not in most cases right okay so um there's a question on whether the prs recovery destination is sequentially updated during a flight whether it's dynamically updated yes uh you mean the uh the

**[30:29]** guided system it doesn't mention um yeah like by updating like our automatic triggering system in our drone is constantly monitoring your flight in real time um we've actually looked into using different uh satellite and 3g 4g technologies to uh to enable um live tracking we did a beyond visual on a site project in 2018 on that um and then as far as the guided system for drone delivery routes there'll be a software that we will be launching but ultimately right now where the the guided system at is is more for package delivery for strategic supplies okay you mentioned that um a parachute recovery system is not just about deploying a parachute when there's a failure how do you establish that communication with the drone when you deploy it and what constitutes yeah so the the parachute recovery system is actually completely independent of the

**[31:32]** drone so if the drone were to fail you still need to have power for your parachute so the way to think of the sequence of events is first is identify failure so that's either with the automatic trigger or it's with a manual remote it's a big red button then you want to kill the power through the flight termination system so again you either interrupt the power between the battery and the drone or you send a command through the flight controller to the motors to stop spinning um for the osdk for example that's kill command and then you deploy the parachute so there's there's a couple secret events to ultimately deploy a parachute system and so for example when we work with drone manufacturers they're directly involved with tuning the auto trigger and giving us feedback supplying us with flight data because ultimately they're going to know their drone and they're going to know their max angles better um than than anyone else and they'll have the data to support it and those are kind of some examples of how we make sure that the auto trigger is as robust as possible okay and what kind of drones do you support uh so right now we've worked on drones

**[32:32]** from one kilogram up to 80 kilograms uh we've worked on vertical takeoff landing uh we've worked on multi-rotors uh right now in market we've got products for the m200 the m300 for the retrofit solutions for you know for the DJI lines we'll be launching the mavic 3 we announced last month we'll have a mavic 3 in market um by the end of q2 that again will support for flight over people compliance and flight termination system requirements and then we work with different drone manufacturers um I can't disclose all of them but you know for example workhorse we did an astm perfect 45 out of 45 with them in april and they're doing things for drone delivery and surveying interesting okay great so I suppose those are all of our questions that we have for today um yeah so uh for anybody who's joining just now if you have any more questions you can visit uh yeah the abs booth or you can email josh as you said

**[33:35]** if you could repeat your email id please for everything yes I'll put it right into the chats my email is josh.ogden at abs and he's also active on linkedin yeah great he's also active on linkedin I believe and we can also they can also connect with you on over on linkedin yes exactly I'm very active on linkedin spamming everybody perfect well thank you very much for having me and look forward to hearing from others great thank you guys yeah thank you so much for joining us