Q & A: Founding CEO at Chironix Daniel Milford in conversation with Defence Connect

Liam Garman:

Well, g'day everyone. It's Liam Garman here, Editor of Defence Connect. Unfortunately, Phil Tarrant, your host, couldn't join us today. I think I say this every week when I say it, and I think, I really need to change that theme tune, that intro, so I get my own one. But you know, perks of management, I guess, for Phil. But today we are going to have a very interesting guest joining us, discussing robotics and technology and AI and all things future, which is very hot at the moment in the industry. It is Daniel Milford, Founder and CEO of Chironix. Daniel, how are you?

Daniel Milford:

I'm great, Liam. Pleasure to be here.

Liam Garman:

Thank you for coming. You have a few very interesting ventures on the horizon, but before we dive into those, I want to get an understanding of Chironix, because you're in a very evolving space. You're in a dynamic innovative space, which when you look at the defence industry as a whole, when you look at science and technology, it's a revolutionary space. Even when you look at the whole economy and the entire industry, it is just changing the way we do things. What does Chironix do, and what do you do at Chironix?

Daniel Milford:

VIEW ALL

Yeah, sure. Look, the purpose for Chironix being formed five years ago was really to achieve one major thing, and that was to deliver robots for humans in such a way that we can work hand in glove with robots that support us in our daily activities. Now, Chironix is actually named after the Greek centaur Chiron, and if you think about a centaur, it's the body of a horse and has the torso of a man. And when we came to really think about how to form the company and what we wanted to stand for and our purpose, vision, mission, et cetera, that's it. It's that sum part of humans and robots being able to naturally interact in such a way that they can actually form a hybrid team.

So our mission really is to build a future where humans and robots work together as a team, where they can communicate naturally in a mutually supportive manner to achieve common goals. That's our mission statement, but it's really focusing on delivering that robots for humans, so supporting people in the field, doing their job with a robot that can actually provide them with greater utility in what they do.

Liam Garman:

Just before this podcast started, we did have a little bit of a chat about that philosophy behind it, and one thing that you said, which I thought was very interesting, it's that you are using the robot technology to get a greater production, greater productivity from what humans are currently doing at the baseline. You add technology into the mix, and now we're just producing and creating a whole lot more at a better efficiency.

Daniel Milford:

Absolutely. Yeah, that's exactly what we do. While Defence is a major vertical for us, we also support the mining, oil and gas, utility sectors as well, and it's the same thing. It's about providing me as a human in the field, let's just say I have a utility value of one, that's how much I can produce as a human in the field, let's just call it one. You give me a robot that moves equipment to me when I need it. If you provide me with access to a robot that can allow me to see or monitor some equipment or look over the side of a knoll and provide me with an observable position information from a sensor it carries, suddenly my utility value suddenly goes up to 1.3, 1.4, because I've been provided with these assets that extend my reach, extend my situational awareness, or provides me with other effects that I don't have to physically achieve on my own. I can slave them off to robot.

It's worth understanding the word "robot." It's actually a Slavic word back from the 1600s, 1700s, which actually means forced servitude, but in the context of animal husbandry. This is literally when you had an ox or a horse in the field and they would drag your plough along. And that context of a robot was actually this thing that directly supported you in doing something. So again, that's where for robots, for humans with Chironix, we're focusing on harnessing this new technology, but in the context of how it can support you doing your daily job to be more effective and more productive.

Liam Garman:

I really like that. That is a brilliant philosophy, because you're just maximising the effort that you've already got in. And we'll get onto this a bit later on, but if businesses take that logic, like what you are saying, just from that base definition point of view of what a robot is, they can really expand their output.

Daniel Milford:

Yeah, totally. Look, this new technology suite that's coming out, really it's just another tool in the toolbox. It's like moving from the Bronze Age to the Iron Age. It's these major industrial revolutions that are happening. And we know what happened with the Luddite movement when the steam engines were introduced into England and the UK. What happened then, you had the Luddites going to actually breaking the machines because of the social disruption. You had the saboteur movement happening in France as well prior to that, when the industrial looms were introduced, because of the social disruption.

What's important with industry 4.0 and the introduction of robots is to control that, to ensure we roll it out the right way. And so what we are doing at Chironix is trying to ensure that we provide the easiest adoption pathway for robots that can provide a consequential service to customers in the field, whereby those members, those employees, can directly use those capabilities to their advantage, whether it enhances their safety or improves their productivity. That's really the secret here. It's actually making sure you can work as easily as possible with a robot into the future.

Liam Garman:

I think you did touch on something there, which is a big thing, especially with the Luddites. We have in modern military ethics as well a lot of people that get sucked into this notion of a Skynet, and it does my head in to no end. You have the consequentialist versus deontologist debate of, well, you know, if with these technologies, perhaps you could have a safer warfare. And then other people on the other side say, yes, we have robots taking lives, and they don't have the ability to formulate opinions, and people losing their lives to non-sentient beings.

And a whole world of military ethics are debating around it. Please, anyone listening, if you want to have a say on that, please do email me, because I can talk about this all day, and I won't take up too much of Daniel's time arguing about it. But Chironix has a very cool product coming out that I want to start off this podcast talking about, and actually, talking about military ethics is probably a very good segue in here, because this is a product which will save lives in the future. This is Project Simpson. Can you please take us through what Project Simpson is?

Daniel Milford:

Yeah, sure. Project Simpson is a programme we've been doing with US Office of Naval Research in support of US Marines, actually for the last three years. We've moved on to a large phase two of the rollout of Project Simpson, but effectively it's using our software and robot stack to provide autonomous loops of logistic supply in the field, with the primary purpose of extracting casualties from war. If you think of it, you've got a combat team out in the field and you've got your brigade headquarters where you've got co-located medical supply or store. The purpose of the robot is to go from that medical depot, travel out to where the casualties or a casualty collection point is. Upon arrival, whatever stores and cargo is on board is to be offloaded by the marines, but then the medical injured casualties are to be loaded onto the platform.

And the platform's designed in such a way that it's specifically designed to take a medical casualty when it's needed. And then they'll autonomously back to the health services node while they're being monitored in real time for their medical status. So there's a number of moving parts, there's a variety of parts, I should say, in Project Simpson. One's the autonomy of the platform, and we're building that software in support of the US Marines. The second is the payload that carries the medical casualty, and that's actually been contracted out to us as well. That's to basically build a really suitable payload that actually can manage the vibrations, frequencies, the changes of the pitch and roll of the vehicle as it's moving across a whole range of terrain, to stabilise that patient.

The third part is to actually then provide the information off the embedded systems within the robot, including the medical systems, and to transport that back to the field hospital in live time, they can see that information. And finally, and most importantly actually, we're writing the operational concept document with a whole range of other international laureates to help really define how we're going to use unmanned ground vehicles in support of military activities over the next 10 years. And there's a big focus on how that's going to be achieved in the Western Pacific theatre. How are you going to be able to achieve island hopping, like was achieved under MacArthur during World War II, with that concept in mind? So we're really looking at a very high level exploration in support of US Marines as they focus on new challenges in the next decade.

Liam Garman:

Just touching on one thing you said about, when you take the battlefield casualty and you move them back to HQ-

Daniel Milford:

Or to the field hospital.

Liam Garman:

... or a medical hospital, to the field hospital, do the marines hook up, or is the idea that the marines would hook up the casualty on vitals so that they can be monitored throughout that way and you can track them, or do they expect that people would go with the UGV itself?

Daniel Milford:

It is a great question, actually. This is where the concept of employment is being worked through with these operational concept documents. So certainly, the marines would actually be strapping on a wearable vital sign monitor and other medical equipment to the casualties as they're embarked upon the vehicle, so that remote monitoring can occur. And that remote monitoring can still occur whether or not they have a local section provided for their own security or whatever security element, or they may not have anything at all.

So some really interesting concepts coming out within the international theatre with regards to, will the allied forces, regardless of who they are, retain air superiority in the future? Will air superiority be in question? If so, when we have mass casualties in a warlike scenario, how do you manage this? And I know there's vehicles of opportunity, which is any vehicle to move a casualty from A to B. There's dedicated AME, there's dedicated field ambulance, the whole raft of ways that you can provide these services.

A lot of is thinking is now being had, and a lot of discussions being had, about what it'll mean in the future when the allied forces, and the US, in this context, comes up against a peer adversary which has amassed a might and a technological capability that may negate some of the advantages the United States has had for a very long time. So a simple question of how they provide local security, if there will be humans around these vehicles, we're developing the technology to allow for that circumstance to occur, where we can outsource the protection to the humans in the field.

Similarly, we're also building this technology to allow it to travel up to speeds of 50, 60, 70 kilometres an hour, to move someone on MSR, a main supply route, from A to B swiftly. And of course, in that context, there may be no humans there whatsoever. The medical monitoring of that casualty is key to that and is key in both situations, and we obviously understand completely that there's a command and tactical decision from the operations groups here with regards to what level of threat they're prepared to accept or prepared to accept in terms of the risk of that casualty being moved from A to B.

If there's a lot of casualties needing to move back from a forward element where there is conflict, and swiftly move them back, there will be a command decision that's just, get them back as quick as possible through as many means as we can. And that allows us to surge autonomous robots in support when that's needed. There may equally be a lower level of warfare in place. It may be some form of counterinsurgency potentially, and there may be a decision being, okay, the loss of the life, the loss of the individual may be reputationally a very high consequence. Therefore, in terms of risk, we want to make sure we've got a man sectioned around it to provide the security. But in regards to moving a heavy person, other payloads, making sure they're medically stabilised during that transit, let's use the robot for the that. So that is very much top of mind about how we provide the command group that flexibility to make good decisions.

Liam Garman:

You've mentioned that obviously this is done in conjunction with the US Marines. You are an Australian business. I don't think we've covered this yet. Project Simpson, two points. Walk us through the name, and do you have any plans to bring this to Australia in the future?

Daniel Milford:

Yeah. Great things with projects is you get naming rights, so Project Simpson named after Jack Simpson and his donkey from Gallipoli, moving patients from the front trenches of Gallipoli back to the beach, because that's what we're doing. It's just not in a beach cliff face. What we're doing is across the entire battlefield, within 20 kilometres of the field hospital location. So yeah, we were quite tickled to be able to name such a large project.

Liam Garman:

Yeah. I mean, that's brilliant. It is quite heartwarming when you think about it, because your tech in the future could be used to save Aussie lives, and it really is the 21st century equivalent of bringing wounded diggers back on a donkey, isn't it?

Daniel Milford:

Yeah, it is. Really is. Again, this is back to our core philosophy that the company is just robots for humans, and it's literally, I just want a robot to follow me, even if it is carrying a casualty as a payload, or to take that human that needs lifesaving intervention back to the field hospital as quickly as possible. Because the issue here is attrition. The entire concept that we're we're playing with is, if we have mass attrition of our forces, we need to make sure that every war fighter available is able to conduct warlike activities, not administrative activities, not these other things which will take them away from the main effort of the fighting force. Back to utility values and productivity, we need to support them as much as possible with these mechanical systems, these mechanical beings, which can support us, because the enemies are doing it as well. So that's where we need to be, at the forefront.

But yeah, certainly tickled pink that Jack Simpson and Project Simpson named after him was picked up as a name, and we're very glad about that. And as an Australian company to be working for the United States government direct as prime, it's something that we've spent a lot of time working through. We had a very good phase one on Project Simpson in 2019, which was incredibly successful. And from that, it's ended up that we've scaled our support of NEXLOG, which is the US Marines group that we're in direct support of, the Office of Naval Research. The budget is tripled for phase two, which is fantastic. Also working with the Naval Postgraduate School and a few of the other war colleges as well, that are providing academic support and insight. And certainly we're growing our relationships with the US Marines positioned throughout the Asia Pacific theatre as well. That's been quite an opportunity. That's been very interesting.

Liam Garman:

And on that, do you see any opportunity of expansion in Australia or teaming with Australian companies to roll this out for the ADF?

Daniel Milford:

Look, we'd love to roll it out to the ADF, and obviously we've made some approaches to the army in particular. I think the army's going through a learning journey about what's right for them. I hope that in the future, there may be some form of interaction there, and the offer's been on the table for some time now. It's really up to the Commonwealth to decide what to do there. However, with Defence Science and Technology, we've had a very good relationship. We've done a significant amount of work with DST and continue to support them with their robot fleets, building bespoke sensors and systems in support of their robots, maintaining their robots, and have also received fairly substantial and continuous funding in development of software that develops the behaviours of these systems.

We're very grateful for that very close relationship we've forged over the last couple of years and very grateful to continue to be able to provide them with services as well. So that DST relationship is strong. I think DST also has the advantage of working very flexibly and seeing what's happening internationally through the Five Eyes network or TDCP as well. But look, we're keen to see how we can work more with the Australians, but we are certainly progressing very quickly with the US, New Zealand, UK, and Canada. We would like to do more here back in Australia, but that's up to the Australians.

Liam Garman:

Well, I suppose it's quite good that you are in a good position, that working with the Marines you're maturing the products a lot more. So by the time you do have Australia on the cards, you do have a very strong offering.

Daniel Milford:

Yeah. Well, the offering's actually strengthening further beyond the military sphere. We are not a sole defence company. We are dual purpose, as you can see logistics movement. What we really provide to our customers is two things. We either move stuff from A to B in support of a human in the field, or we monitor stuff from a variety of locations and remote operations in that context. And because of that and our location being headquartered in Perth, means that the mining sector, the oil and gas sector, the utility sector are key verticals that are well ahead, well ahead with their adoption of robotics. They understand interoperability. They understand a lot of the challenges with infrastructure.

We've been supporting some of the biggest companies in Australia for many years now, and we're very glad for the work that we've been doing with them, because that technology investment they're placing in us provides great opportunity for our other defence consumers and the products we're developing there, and vice versa. With our work with defence, we build novel and niche products, which are more applicable in the battle space. But then suddenly if we have to do something down a mine shaft where GPS isn't available and these other technologies need to come to pass, or if we need to suddenly detect noxious gas in an oil and gas plant, and all of things we've done, then our investment from our defence customers allows for that product and technology be immediately available to these other national actors.

So I think what's key here is, if the Parliament and the government truly wants sovereign capability and there's this drive to build indigenous industry here in Australia, we're very keen to provide that. We've got good commercial support. We've got great international support. I'm still wanting to work much more with the Australian government in how we can provide it much more of a total service. But I certainly can say from the industry side of our commercial partners in mining, oil and gas, utilities, et cetera, they are scaling and they are moving and they are reading the winds of change and are moving ahead. I know that Defence is a big beast. I was a former public servant for six years myself, and I worked in capability development, so I know how challenging things can be, but change is upon us, whether we like it or not. And it's the time to adapt is now.

Liam Garman:

And that's a sentiment echoed, I know, especially in the space industry as well, when you talk about just how surprisingly short the jump is from Australia's mining industry into defence. One thing we always hear from space providers as well, and this ties straight into autonomous and uncrewed vehicles, space providers working with the movement of autonomous vehicles throughout large mine sites and coordinating the movement of those vehicles. And now they're working with those and the application of that technology into the defence space. It's surprisingly quite a small jump, as you've identified.

And as we move into the future of more technologically advanced warfare, we will all be moving into more teaming systems, whether with crewed and uncrewed systems together or just uncrewed systems together. Like you mentioned before, centres making sure that they can work together, not clash, not collide, and team in a way that you can really dominate the battle space. It is a growing field and a very small jump between your commercial clients and Australia's primary industries and defence.

But on that note, we're going to have to go to a quick break, but when we are back, we're going to dive into that a little bit more and learn, discuss a little bit more, about applying these robotics in the commercial space. So we'll be right back in just a second.

And we're back, and I'm joined by Daniel Milford, Founder and CEO of Chironix. For those of you who were with us before, we just discussed Project Simpson, which is Chironix's new battlefield casualty system, which will save lives once fully incorporated into the defence structure. It brings back casualties back to field hospitals, and it's quite an amazing innovation there. But Daniel, I saw some news that Chironix put out recently where I think it was quite direct. It was a very direct suggestion, I suppose, to businesses that if you don't embrace AI and robotics, quite frankly, you're going to fall behind. You're not going to be competitive. What's with that? How far into the future are we, how much are things changing, and are Australian businesses really at risk of being less competitive?

Daniel Milford:

It's a great question. And yes, it was designed to be a provocative article, to have the opportunity and to raise the opportunity for some of these discussions to be had. Chironix has been in the unique position over the last five years of our journey, that we work with international partners like Boston Dynamics, Clearpath Robotics, Agile, Canadensis, all of these other manufacturers. And as you understand the global industry and the adoption of all of these different technologies and you realise globalisation is a real thing and supply chains, and you even see where our batteries and everything come from, you end up conversing with some real luminaries of industry, people that are so close to this technology and where it's been deployed, that you get a very deep understanding of adoption and you get a very deep understanding of outputs and the outcomes of that work.

And something that's very commonly put to me by these other partners that we have around the world is that we're seeing this growth in robotics in the US and Europe and Middle East, and there's mass adoption, and there's all these benefits coming out, and there's changes coming out as well, and they're pivoting and adapting to it. And we're not seeing anything really happening in Australia and New Zealand much. What's going on? Why haven't we seen this growth? I have these conversations about it with a whole range of customers. I think sometimes we forget a few things, that we are in a globalised economy, where we are in such a safe nation as we are in Australia, in such an interesting part of the world, that we often just don't have that same cultural impact that you would like with someone in France, where you've bordered with all these other nations in their own journeys, and you can have all that cultural exchange. Or like United States, that's pushing so hard on the technology edge.

So there is a lot of change happening overseas. Other nations and luminaries would say that we're behind in our adoption of new technologies, which if we're moving again from the Bronze Age to the Iron Age, means that we're not as competitive. And if competition and achieving what our interests are for the next decades in a globalised economy, we need to really think about how we're going to do that. We need to think and realise that with the new reports which are coming out, talking about intergenerational change, so the Intergenerational Report 2035 that came out, it was talking about the fact that we are going to live a life which is of less quality than we are now and what our parents have experienced, because we don't have enough people in the workforce.

Our population's just not expanding as high as it needs to be expanding to support our way of life. Ways to get around that is automation. Ways to improve productivity in our workplaces and enhancing safety is through the adoption of automation. There are a number of challenges a nation faces, and that business leaders face, which can be overcome with technology which is now available. It's just depending upon their maturity and their ability to want to undertake technological change. And this innovation is part of it. And "innovation" is a bit of a dirty word as well. It's copped a lot of flack over time. It's really just taking emerging technologies and applying them to the workplace in a way that's going be successful and demonstrated to be. That product has now come.

Liam Garman:

That ties into exactly what you were talking about before with your verticals. You work in oil and gas as well. Can you provide a case study there of where this has really made a big impact in Australia?

Daniel Milford:

Oh, look, absolutely. Yeah. A very simple ones is remote monitoring. Let's just use a few blended use cases here. Robots on platforms at sea, which are unmanned, which you are able to remote into, turn them on and travel them around in a patrolled loop around the unmanned station at sea, so you can look at any of the gauges or readings of the sensors that the robots carry. Very simple in its context, but the impact is enormous, because no longer are you spending $150,000 to spin up a helicopter to take a crew of people from the shore to the station, just to see that it's actually a false reading. That's a very simple one.

Again, mining sector, robots that have been used for cave exploration, robots that have been used to do some monitoring on sites and areas, monitoring that are doing photogrammetry analysis of the subsidence of the pit walls, things that humans aren't doing regularly enough to ensure safety. But when those things are not done and we lose lives, there's a big question of, well, why aren't we using this technology we have available? Because if we had done this in the first place, would someone be alive today? And despite the tragedy of a death, there's also the cost. Immediately the moment this situation occurs, you would've paid off the robot five, six times fold. So that is some very simple ones.

If we look at the Project Simpson context, which is logistics in nature, again, if we've got miners in the field, usually the drill teams, and we want to move different pieces of equipment to them, or there's a broken down truck in the field, you no longer have to drive from where you are all the way back up to the warehouse or the maintenance facility, which is a 40, 50 minute drive out of a mine site, pick it up, whatever you need, and drive all the way back out, so it's an hour and a half gone from your day already. Instead, you just make a phone call. They load this on the robot. It's loaded on a robot. It comes out with only a 30 minute drive, and suddenly everything you need is there. So you've haled the downtime immediately, and that's important.

When you think about the mining sector, what you really are needing to consider is the fact that it's a production line. It's a production line from everywhere where they identify the product, blow up the product, crush the product, stick it on a train, get it to port, ship it to the customer. That's their production line. Any disruption to that is an economic impact, and they want to that eliminate that. Defence does have some lessons they can learn here. There's a lot of interoperability. Equally, defence does some great things in the war fighting context, in terms of the process and the procedure, that other industries can benefit from as well.

Liam Garman:

That was going to be my question. We do have a lot of SMEs, Australian businesses, working the defence industry that do listen to this podcast, trying to get a little bit of insight into how they can... well, apparently "innovation" is a dirty word, so I'm going to try and not say that anymore... make better how they can improve their processes and procedures. In this sense, you said mining is a production line. A lot of defence businesses are quite literally production lines, as they're supporting a sovereign industry.

And I feel like there's a little bit of feedback in the industry that we have this very ethereal, nebulous notion of, oh, we always talk about sovereign all the time, but no one ever really defines it. It never really goes anywhere. It's a bit of buzzword. In a defence aspect, how can automation help? I don't know if you're allowed to give cases, because I know we're all usually about 15 nondisclosure agreements in deep, whenever we work with people, so you might not be able to give a case study, but how can these businesses innovate with an uptake in automation?

Daniel Milford:

Okay. Let me try and generalise. I'll put it specifically, I'll answer that specifically with relation to defence industry, because I can talk about a whole range of other sectors and benefits and improvements, but if we look at shipbuilding, for instance, there's a couple of things you need to achieve or any workplace needs to achieve as you go through the process. Particularly in construction and building, inclusive of shipbuilding, is monitoring the status of the build. And there's a number of sensors people are using, regardless of robots, but a number of sensors people are using to monitor and track the build state. That can be LIDAR point cloud information. It can be a range of cameras. There's a whole range of service providers which provide those cameras. And that's about understanding what the build state is at this point in time from this position.

And then basically matching that with other computer-aided diagrams or other information, so they understand schedule, where they're at and what are any of the issues. Similarly with any major facility, that's a vital asset. You also then have the layers of security and protection, and you're trying to understand builder analysis. You're trying to understand what has changed. Tell me what has changed. And so you set up autonomous patrol routes that have cameras again on board, you can detect what has changed in the fences and other areas.

So first of all, you're achieving enhancing your security awareness of your perimeter in the latter case, but in the former case, what you actually are doing, you're providing more data so that you don't miss things. Or if you do find an early fault, you do find it early in the process. You don't find it later at the end when you're trying to put things together and realise you've got a one foot gap between this assemblage and that assemblage.

Liam Garman:

Once you've assembled your Ikea furniture and you've got the three bolts left over, and you're like, oh no, where did this come from?

Daniel Milford:

Exactly, yeah. But if you think about that as the difference when you're putting together major ship components, whole groups, then that's a significant setback. So it's trying to eliminate rework in many ways. Much comes down to the sensor. If you think about autonomy with the context in which we operate in Chironix, we have a mobile ground platform, and that ground platform can carry something for a period of time. Our smaller robots carry generally sensors that are useful to the customer, and our larger platforms carry kit and stuff. That's really it. But how that impacts the workplace is substantially variable between each of our customer bases.

Outside of Chironix, or to automation in general, obviously you've got production line with robotic arms. That's not what we do as a company, but that's one way of automation. Automation in terms of software, automation in terms of how you go through a process and basically enhance the way you provide a service. So that's any of the productivity means you can do. I think what's important, however, with all this conversation to understand is that we as business leaders, it's important that we understand that we're not in an isolated condition. The nation is not in an isolated condition. We're in at least a decade of competition, and a lot of the safety and security, even in terms of just business security, that we've experienced historically throughout the '90s and '00s and the growth of all these sectors we've had is starting to diminish. And there's a lot of opportunity, but there's a lot of competition.

And so sovereign capability from a government's point of view means resilience, capable, innovative companies that can support the national agenda in a time of crisis, and we've got a lot of way to go. We didn't manufacture vehicle tyres up until last year in Australia. There are some really critical... COVID has been fantastic for industry in a very destructive and challenging way, because it's demonstrated how brittle some of our supply lines are, and the supply chain dominance that is held by other nations, not ours. Even with our major acquisitions for these big major projects we undergo, it's interesting to see how many major foreign primes dominates our national agenda.

Liam Garman:

I do like that point about COVID. It was a great wake-up call. A few weeks ago we had the Defence Connect AIC Summit in Canberra, where we got some key government decision-makers together, and they were giving their feedback on a lot of applications by defence companies who are trying to break into the defence supply chain, defence ecosystem, and talking about this era of competition in a business aspect. It was kind of surprising how many businesses didn't get the basics right. They basically said, some of our panellists said, you know, you can have the coolest, amazing, most innovative product, but if you source it from the one location, well, your liability for defence.

If you don't have very strong cybersecurity protocols, well, there are just going to be downstream risks to everyone else in the supply chain. You're not getting a look-in. If your staff don't have security clearances, you're not getting a look-in. And there are a lot of very basic things that a lot of companies aren't doing. I know that's a little bit of a tangent, but it does go to show that we do work and we do live and operate in a very competitive field, and you do need to be better than your competitors to actually break through.

Daniel Milford:

But therein lies a big challenge. The DISP, the Defence Industry Security Programme, is great. We've gone through our accreditations, and we provide those services, and it is good that it sets a bar, a hurdle. It says that you need to achieve this in order to be cleared at this level, and that's useful. It's really easy for SMEs, however. If I put it in that guise, we make an investment into these systems and programmes to ensure we meet the requirements of the department, and if that's not matched with acquisitions and if that's not managed with some form of revenue generation to offset those costs, then there are questions about how to work more effectively in the defence market.

Liam Garman:

So those frameworks provided by Defence are the best way to operate in the defence industry. Do you think they're sufficient in building a more robust workforce in Australia now that we're talking about creating sovereign supply chain?

Daniel Milford:

Well, look, I certainly think that they actually establish parameters upon which the department expects to be met in order to work with the department at certain levels of security. Obviously, the Defence Industry Security Programme is one of these key ones. It does set out what compliance means. I think the challenge, particularly for small and medium enterprises in Australia, however, is that there's an investment made by companies and by small companies here to match the requirements of DISP. However, at the same time, once you achieve that level of status, it's then matching the amount of work that you get through to offset the costs that you've made to ensure that you can work at that level.

So I think, yeah, it is very interesting, because what we are really talking about is acquisition here, and the investments that companies make to work with Defence, they need to offset the costs that we have to match in order to work with Defence. We have to meet these levels of acceptability, so therefore we do expect if we meet the levels to be given more work. And I think that's been one of the challenges of the department for a while. There are all these tenders with major acquisitions and foreign primes dominating that and then subcontracting all of the work to a whole variety of actors. That's challenging. Trying to have some awareness and transparency on acquisitions across the department, because there's so many silos, and particularly in these entrepreneurial innovative product areas. There could be spurious acquisitions, a whole bunch of ways of going through it.

So it's difficult to understand it sometimes what the department's priorities are, apart from generally being traumatic. So it's challenging, but I certainly do think, at least now Defence has specified you need to meet these requirements to work at these heightened level of securities, but there is a business expectation once those security levels are met that there should be work generated as a consequence of achieving this. And after talking to a number of other small business owners, this may not be my personal position, but after hearing the things from them all, there is this element of perhaps those expectations not being met.

Liam Garman:

Yeah. Yeah. I think it's a great area of debate amongst defence SMEs about the direction given by Defence to operate in that supply chain, that ecosystem.

Daniel Milford:

There's nothing greater than supporting the national agenda. We're all very proud of it. We know there's nothing better than your piece of equipment or software in the hands of a military member, and you know that you are providing in effect to them. We all strive for that. I think the US, patriotism is what they do. I think for Australia, we sit back quietly, give ourselves a pat on the back and have a beer and go, yeah, good. We did good today. We saved someone's life.

It's important that if we are going to support the national sovereign initiative here, that we see those wins, however, because that's what supports and grows jobs. And if we aren't growing jobs in Australia for sovereign capability and we're growing them offshore, then what are we doing? What do you mean by sovereign capability? If we're just going to import technology all the time and not focus on key areas of development, then that's not sovereign capability, that's just an acquisition programme where you just buy stuff from overseas and potentially manufacture some of it here, but all the intellectual property is overseas.

Liam Garman:

It's not a 100% sovereign supply chain.

Daniel Milford:

It's not, no.

Liam Garman:

And on that note, we're going to have to go to a quick break, but when we're back, we're going to have the challenging round, where I'm going to unfortunately ask some hairy questions of Daniel about unmanned maritime vessels. And we're back in just a moment.

And we're back. And for those who were listening to us before the break, we did a quick wrap-up of Chironix's Project Simpson and some very strong advice from our guest Daniel Milford on the need to innovate for Australian businesses. But Daniel, I have a question, because we were talking before about your work with oil rigs and using uncrewed vessels to perhaps check a sensor that's flagging an issue, instead of having to get a helicopter out there, flying a crew out, checking out, potentially at huge costs to the provider. You can get a robot to do it for you. And that's the philosophy of Chironix, to really build human productivity, which we're kind of lacking in Australia, I think is the consensus.

But I'm going to throw a bit of a hairy question to you to wrap up the podcast today, because I was reading an article perhaps throwing a little bit of shade, a little bit of oversight on large scale ships, especially warships, the notion that they're becoming slightly more uncrewed. They're trying to reduce the over-reliance on having a lot of sailors, obviously, there's a human cost to it, there's a staffing cost, with the idea that having more uncrewed or less crewed ships will be better in the future.

However, one bit of feedback and some oversight on that was that some issues are allowed to fester over time, get worse, because there are fewer sailors on board, fewer crew on board to catch out issues as they arise, and undergo maintenance while at sea. Being an expert in this field, and I'm not suggesting you're a big shipbuilder building these, what would your feedback be to that? Because I know there's a lot of discussion around moving to uncrewed maritime vessels, and I'd love to get your insight into that.

Daniel Milford:

Yeah. This is a big space, because obviously our major ships that we acquire in the government, they have a life of time for a significant period of time, 20, 30, 40, 60 years. And it's understanding during that time, they go through a whole range of upgrades. I think many of us listening will be familiar with fitted for, not with, a lot of the concepts of buying hulls and putting them in place to actually then improve the technology over time. And it does come to a point that we're introducing more autonomous solutions on an existing hull, that it really does change the way that the vessel is maintained. It changes the way, intrinsically, how the vessel operates.

And it's good that there's actually a lot of experimentation happening with this, because shipbuilders and the requirements generation groups within the department have the opportunity to ideate on what's necessary for the future. It's interesting to understand what the purpose is of a vessel at sea and what the effect is that you want to generate, and to start there. Look, I'm very aware of the Sea Hunter programmes and what's happening in the United States with the US Navy's autonomous submarine tracking vessels. The purpose there really is to monitor the sea below the vessel and to basically broadcast any change, so it's an appropriate use of that platform. It's not doing counter-piracy. It's not doing logistics provision. It's not doing demonstration of capability or firepower and diplomatic or influential role. It has a very specific, dedicated task, which is submarine tracking.

It's important to understand where autonomy is useful and where it is not. When we in Chironix think about robots for humans, we understand there's two circles in the Venn diagram. One of the circles is humans and our capabilities, the [inaudible 00:43:22] cognitive capabilities in our sapien nature. And the other side is robots. Very repetitive. Very much has to be a known environment and iterate on exactly the same thing time and time again. But they can carry significant payload, do basic autonomous patrols and solutions repetitively without getting bored, and the four D's play there: dirty, dull, dangerous, dear tasks.

So it's understanding how to leverage both of them to provide the effect that you want to achieve. I do think with regards to autonomous vessels at sea, you just need to also understand what the effect is you're trying to generate. Starting with an old hull, something you're trying to modernise, having it optionally crewed or crewed with some members, some of those work tasks need to be really figured out of what needs to be done. You're going to need to change filtration units. You're going to need to do maintenance and inspection. Now, you're still going to need a crew on board to ensure the vessel is sailing effectively. However, if you automate a whole range of the other features of the vessel, you need to consider the impact upon the operable capability of the vessel.

So yeah, it is a very complex area. Obviously, I'm quite aware that DARPA has been doing a lot of research in this area, and there are some decisions that need to be made through this learning, which is actually what's really happening here, about what are vessels and how are they going to perform in the future? How should they be designed? Should they be designed with lots of cabins or just a few? Will we have humans on board? What are the implications for remote control in operation of them, in terms of someone can effectively take over control? Does that mean we always must have people on board the ships? A very complex area. Lot of questions that need to be asked. At least I'm hopeful that by doing this experimentation really, that the problems are becoming very clear and we can start solving them.

Liam Garman:

You'll know what works, what doesn't work, what performed better than expected, what performed worse than expected. Extrapolate from that for the future. And I suppose there is intelligence in saying that some things for now remain in that Venn diagram, like you were saying, remain very firmly in the human sphere. You can't set a weapon system to fully autonomous. That would most likely break the laws of armed conflict, and we've had it before, not with fully autonomous weapon systems, but with passenger jets accidentally being shot out of the air. You can't set something to fully autonomous. That's something that would stay very firmly in the human sphere. And maybe even things like changing filtration system, those tasks might stay as well. And we would learn through iterations and through repeated trials what we can say, oh, that worked really well, that stays firmly in the robot sphere. And then from that, we build and expand and learn and innovate more.

Daniel Milford:

Yeah, absolutely. That's all necessary. This is where you need to understand the capabilities and dynamism that you get from a human versus an autonomous solution. And look, the next 25, 30 years is going to be extraordinarily challenging. You mentioned international humanitarian law and laws of armed conflict and laws of engagements, and what's going to be appropriate within an OODA loop even, considering the observer is automatic, where a targeting system has to actually be engaging a legal or a human to determine whether or not to actually engage in trying to take a firing system. I think some of these ethical moral boundaries are going to be really challenged over the 30 years if there is significant conflict, because the Western way of war and the nature and character of war that we have as of today is not held cognitively to be the same globally. Other actors may consider that international humanitarian law type of perspective to be different. It's a very challenging area.

Liam Garman:

With the OODA loop, it's funny you should bring that up. One thing that I was taught was that technology, in an air force circumstance, not that I was in the air force, by the way, but technology enables pilots to shorten their OODA loop, but the OODA loop still exists. The decision realm still falls within the human matrix. It's not falling under the robot matrix, but the robot matrix shortens that OODA loop, so you can undertake a decision quicker.

Daniel Milford:

Yeah, that's right. That's the autonomy, less the robot, the autonomy side. Sorry, an interesting glossary in the language. The robot would be a physical being or actor in the environment, and the autonomy can be the software and everything else to support it. But certainly where those things are active in the environment and they shorten the OODA loop and provide better quality information to the user, we have to actually then match it in many ways and make more effective and quick decisions. It's not just a challenge of these autonomous solutions having the capability to actually undertake a whole range of independent action. It's also understanding where we need for humans legally, morally, ethically to be involved in those final decisions.

But I think there's going to be great challenge with information consumption on a human behalf. I certainly find even with our simple deployments of robots in support of payload manoeuvre or monitoring, we need to be very careful about the level of information we provide and making sure it's not data. It has to be information or knowledge to the end consumer, so that they can get on with their daily activity, and when it's necessary for them to be informed about something that's changed in the environment or something they need to be aware of, we give them that information at the right time to change their behaviour and what they're about to do about it.

So it's very broad. I know that's a simplified use case, but I think that's important, because as we start to consider the complexity of a hull, all of the subsystems on it, even automating the dispensing of food or water or drink, all the way through to a targeting missile system, it is extraordinarily complex. Automation is available across the entire platform. I think it's often worth just simplifying what you're trying to achieve as an effect. And certainly within our area, we focus on the four D's: dirty, dull, dangerous, or dear tasks. Any of those tasks which we can affect with our tech stack, with our capabilities and systems, that's what we focus on improving, because we know it's going to get a net benefit to the customer at the end of the day.

Liam Garman:

And one thing that I always have a huge issue with, just a bugbear, it's just those things that just really make you angry, is the blurring of the lines between autonomous and uncrewed, to a point now where a lot of people, not necessarily in the defence industry, but a lot of people generally, see them as synonymous. And they're very not synonymous things. An uncrewed system and an autonomous system are quite different. And one thing that I think we're facing in the defence industry as well, you have a lot of humanitarian groups and a lot of advocacy groups coming up.

I think one of them is called Stop Killer Robots, and they're really anti-drone systems, because they're suggesting that in a combat environment, the decision to kill is being taken out of the hands of a human or put into a Skynet robot algorithm. That's not true. That is not true, and they've kind of made autonomy and uncrewed this synonym, and they've blurred the lines between them. I think in the grand debate surrounding innovation in defence, it's not necessarily helped too much.

Daniel Milford:

No. And you know what? Defence isn't a good communicator either, at the same time. Sorry about bringing up Clausewitz at the moment, but the trinity is there and we need to bring the population that pays the bills here along on a journey, because we don't want to be seen as introducing Terminator into a civilian space where we're affecting another population and lose the moral authority to act on behalf of our nation state. So absolutely, communication is key here, and frankly I think it's the job of the government to make sure the public understands what it stands for and what level of force and how we will provide that force to areas that we're trying to help, really, at the end of the day.

Levels of autonomy, as we are talking in industry, we're talking about how the platform's getting from A to B, how it's navigating, how it's experiencing the environment and making decisions about moving left or right. But the decision about whether or not to release a missile or the decision about whether or not to identify that target as potential enemy, or at least non-friendly, it's communicating and educating people that that is a human process. Defence, it's really there for the purposes of war. This is a war process or a military process that's been going through. There's always going to be a human element there.

So there needs to be a level of awareness and learning. And of course, there'll always be actors which are just not really understanding the purpose of defence. They're a minority against it. But yeah, their communication is really key and important. And it's worth also understanding here that robots are being deployed in the battle space beyond the UAV out-of-human-site arena, as we've seen the last 30 years or so. Robots in close proximity to humans, it's just a new phenomenon, and particularly where probably in the last 10 years, they've only started being introduced in the battle space. It's going to be very interesting if we are trying to win hearts and minds about the message that they will convey.

So we've been deliberate within Chironix to date to ensure that we're focusing on dual purpose capabilities, manoeuvre, navigation capabilities, observance, all those things, because we know economically that's our best opportunity as well. And look, there are, there are defence primes that specialise in providing Defence with a whole range of munition services and so forth. There's no competition there that I want to partake in, just economically. But yeah, certainly communicating with the public, making sure that they understand what autonomy means in terms of the context in which we're discussing, it's not the kill chain, it's autonomy with how it gets from A to B, that's important.

And look, I think the George Lucas effect is what I talk to clients and staff about regularly, keep you in space with lasers making noise, it's just an understanding of basic physics as well. So certainly what you see in the media about autonomy and these amazing marketing campaigns that they put in, there's nowhere near where one's realistic, and the fragility of an autonomous system, which can be there. But certainly I think it's important to shape the conversation and to shape the outcome of technology. I think the final statement there is that the technologies which ultimately we will adopt are the ones that are invested in. So if we want to invest in autonomous navigation on a whole variety of platforms to achieve a supportive effect for the human workers that are relying upon that effect, that's brilliant.

If we want to focus that funding towards autonomous kill chain solutions, to speed up our ability to affect a near peer aggressor, then that's also an equally viable solution. It's just whatever gets the funding will become reality. And as there are military ethicists, there are military strategists that are out there that are thinking about this technology, and Strategic Policy Division in Defence and a whole bunch of other military strategic commitments, if they're still called that, need to be considering what they need in the future and the way that they want to effect the likely plans that we have for the next 20 to 30 years as a nation to provide government with a relevant operational capability that matches that future state.

So yeah, interesting, very challenging area. A lot of robust and really good discussion should be had more about what we want from autonomous solutions, and that transparency of communication is key to bringing the Australian public along with the Defence idea of what developing the sovereign capability is really going to be.

Liam Garman:

I think the legal and ethical argument can make a whole podcast by itself, so I might need to have you-

Daniel Milford:

Oh, absolutely.

Liam Garman:

We might need to have you back in a few weeks because no, you're totally right, some of the stuff that you see in the media is completely fantastical about these capabilities. And actually a lot of it's just downright not informative of what... Project Simpson is an amazing example of a capability which in the future could save lives, save Australian lives. And it's a smart system which is built for great intentions and for great outcomes.

Daniel Milford, thank you so much for joining us. I think that's probably all we have time for today. Daniel, thank you. Very informative. Very informative discussion. I think a lot of our listeners are going to be left especially with that last little wrap-up, the ethical, legal wrap-up. I think we've given people a lot to think of, but thank you so much for joining us today.

Daniel Milford:

Thanks, Liam. Great to be with you.

Liam Garman:

And to all of our listeners, thank you so much for listening to us, and we will be back at the same time next week. And as all always, goodbye.