While the RAAF is contemplating delivery later this decade of Triton Unmanned Aerial Systems (UAS) boasting wingspans of nearly 40 metres, the DSTO is meanwhile studying the complex interaction of air flow over the infinitely smaller wings of insects
The two are connected – the DSTO research is intended to contribute to the development of micro air vehicles (MAV) that could utilise flapping wings for lift, propulsion and stability.
This work is part of what Dr Ken Anderson, Chief of DSTO’s Aerospace Division, acknowledges is a fairly fragmented DSTO strategic research program into UAS systems.
“What we have at the moment is a relatively small group of tasks which aren’t particularly unified. We’ve got a number of relevant issues that give us some insight into what’s happening in the unmanned aircraft world; we’re putting ourselves in a position where we could take on a bigger project if required to do so, to be well across the technology challenges and to have a basis for relationships with other parties,” he said.
Flapping wing technology
In the case of flapping wing technology, the other party is the zoology department of the UK’s Oxford University, which is heading a substantial research program into insect flight that is supported by that country’s Engineering and Physical Sciences Research Council.
“As always, we participate in a small way to get some insight into the larger program,” Dr Anderson says.
“We’ve done some work on trying to understand the interaction between flapping wings; in effect how to understand the airflow and particularly how to model it.
“It’s particularly difficult when you’ve got two sets of flapping wings as you do with a dragonfly; when you see the wings flapping in sync that’s probably because it’s moving forward and generating some thrust whereas if they get out of phase it’s probably hovering. That’s the way an insect controls its speed as opposed to its lift.”
With fixed wing and rotary platforms losing efficiency the smaller they are, so-called flapping flight has obvious attractions for ISR tasking in urban and confined spaces, if and when the concept is mastered.
This bio-inspired and other UAS research is included in DSTO’s long-term strategic research plan. Unlike the majority of the organisation’s work, the plan is self rather than client-generated and outcome focused, with the objective of providing insights into areas of possible future importance.
Such long-range research is always vulnerable at times of fiscal stringency, but Dr Anderson says that without it, DSTO would morph into a short-term problem-solving, reactive organisation – “We would essentially just be applying the knowledge we acquired yesterday.
“A lot of our UAS work is bottoms-up ideas and that’s what happens with a discretionary strategic research program, lots of little ideas.
“At some point though it’s prudent to take stock, think about the bigger challenges and then try to develop and integrated program, and that’s where we are now.”
A range of challenges
Part of the challenge for DSTO In deciding on where to position its UAS research has been a surge of interest in UAS technology, boosted by the low cost of entry.
As acknowledged by Dr Anderson, the current focus is therefore on disparate issues – “There was no single glaring research gap - on many of which DSTO is collaborating with international partners and the university sector.
“There’s so much happening in the civil world it’s not enough just to collaborate with our colleagues in the UK, US, Canada and New Zealand via the Technical Cooperation Program. A lot of universities are working on things like autopilots, flight control, and how to operate more effectively in urban spaces where there are problems such as wind gusts.
“Many of the civil applications are relevant to defence too, so we’re very interested in knowing about the work some other people are doing.
“In the main the programs are led by the relevant university and we provide support; sometimes money, sometimes equipment on loan, sometimes access to our test facilities. It’s good value for them and good value for us.”
Within Aerospace Division, an internal theoretical study is exploring the extent to which an autonomous system might be able to make its own job decisions and operate without direct control. This raises questions involving confluences of design and data, together with the consequences of things going wrong.
“Many of what we call autonomous systems are far from that, they’re acting on very precise and controlled conditions so the idea is how we might develop a rationale for allowing things to be more autonomous,” Dr Anderson comments.
Collaborative research
Another collaborative program is researching the incorporation into autopilot design of optical flow sensing for navigation; an issue of particular interest to small, low-flying UAS in urban areas where GPS communication can be unreliable and the aircraft is vulnerable to obstacles, especially if slightly off course.
Safety in a wider sense is the subject of a planned paper study on UAS risk, including airworthiness classification.
This will take into account the fact that UAS are often built to different airworthiness standards, particularly if they’re to be operated in remote areas with little or no risk to people on the ground.
“When people want to use them somewhere else there will be different regulations to be satisfied, and we want to get some experience on what new certification regulations might be based. It’s particularly true with sense and avoid,” Dr Anderson notes.
“If a very cheap unmanned system operates in an unpopulated area and hits an obstacle like a power pole or a tree you just take the crash.
“But the more you invest in terms of smart sensors, smarter navigation systems, and smarter data transmission systems, you probably don’t want to lose the platform, so having some onboard obstacle-avoidance systems would be a good idea.
“The work here is just to try to understand risk a little better with a view to talking to the airworthiness authorities at some future point.”
Discussions on the subject have taken place in workshops with technical counterparts from other countries, more to establish the value of a sense and avoid scheme than to plan its design and development.
“The more you put requirements like that on an aircraft the more the cost goes up and so does the payload weight. That will affect endurance, and the issue is whether it will really reduce risk in a meaningful way or simply mean more overhead.”
A further line of collaborative research is aimed at increasing the endurance of electrically powered UAVs. These aircraft tend to be cheaper and smaller than platforms powered by hydrocarbon fuels and are certainly quieter, a significant tactical advantage, but endurance is no more than two hours and sometimes as little as 30 minutes.
Lightweight fuel cells
The first step, according to Dr Anderson, is to use fuel cells to help recharge the batteries, in parallel with solar panels fitted on wings that would need to be fairly long and thin.
A lightweight fuel cell can supply the relatively low levels of power needed during cruising and loitering, while the battery is used during take-off and climbing, and is then recharged by the fuel cell.
Experiments and modelling have been carried out in collaboration with the University of Sydney to characterise the performance of hybrid fuel-cell/battery systems on small aircraft conducting surveillance missions.
DSTO scientists and their counterparts at New Zealand’s Defence Technology Agency (DTA) are also preparing for a flight trial aimed at demonstrating the integration of a flight-worthy fuel-cell system on an existing small UAV.
DSTO’s modelling has shown that replacing the battery pack in DTA’s 2.3metre wingspan Kahu UAV with a fuel-cell system could double its range and endurance. Aircraft designed specifically to incorporate a fuel-cell system could see dramatic improvements.
Solar harvesting that could provide UAVs with limitless energy is being studied with the Australian National University (ANU) using lightweight, highly efficient Sliver photovoltaic solar cells.
DSTO’s modelling indicates that despite the extra mass represented by solar cells and the accompanying components, a net benefit to aircraft range and endurance can be expected even on partly cloudy days. A flight trial at ANU of an aircraft carrying Sliver cells is anticipated in coming months.
“These things can help, but the really interesting question is whether we can make a UAV soar like a glider; in other words seek out the thermals and circle upwards, then take off again in its chosen direction to another thermal and climb up again and proceed,” Dr Anderson states.
Control algorithms that permit an aircraft to identify and autonomously exploit thermal updrafts have been tested through high-fidelity flight simulations and in a flight trial conducted by DTA and DSTO.
Although the benefits of thermal soaring vary with local weather conditions, DSTO’s modelling has shown that the range and endurance of a small UAV can be increased, on a mean, yearly basis, by a factor of ~5, for regions with relatively favourable climatic conditions. For an aircraft with a nominal battery life of two hours, that could mean 10 hours of operating time.
A DSTO/DTA flight trial conducted with the Kahu UAV in New Zealand in 2010 demonstrated that autonomous soaring guidance and control could permit a retrofitted UAV to gain altitude with its electric motor turned off. Only a small change to the UAV’s control software was required.
DSTO scientists have also recently begun collaborative research with a team at RMIT University that aims to be the first in the world to demonstrate autonomous-control techniques that would permit a small UAV to remain airborne using updrafts around buildings. RMIT students and researchers will conduct a flight trial with a small glider in coming months.
