Virtual panel discussion: Drones

November 12th, 2015, Published in Articles: EngineerIT

Drones are a promising part how we manage our world. They will be the intelligent connection between people, process, data, and things to create new value and opportunities. We asked a number of experts to share their views about various aspects of the drone industry and the application of drones.

Mark van Wyk, FlyH2 Aerospace; Duran de Villiers, SteadiDrone;  Hannes van der Walt,  The Gadget Shop, Rolf Schlub, TerraCam and Peter Cosyn, Gatewing.

Mark van Wyk, FlyH2 Aerospace; Duran de Villiers, SteadiDrone; Hannes van der Walt,
The Gadget Shop; Rolf Schlub, TerraCam and Peter Cosyn, Gatewing.

With the drone market increasing in diversity the question arises around the design and development aspects. What do you believe is the biggest challenge to the design and manufacture of a universal platform, adaptable for a diversity of applications?

Mark van Wyk, founder of FlyH2 Aerospace does not believe there is a single one-size-fits-all platform. “I believe that these aircraft should remain specifically designed for their application. There are always differences in an application. For example, the survey aircraft we’re developing will never make a good quick response aircraft as there are compromises we had to make during the design. We used long wings to ensure low power flight resulting in long endurance and heavy payload carrying capability. This means that our aircraft will not work well on a rail launched system, and requires longer assembly and disassembly. We chose to use hydrogen fuel cells as a power system. This allows long endurance, dead-silent operation, but doesn’t allow for high power, high speed flight and steep climb rates. This being said, there are some amazing achievements being made with multi-rotor designs. There is a company claiming to be able to build a multi-rotor aircraft capable of carrying 3kg payload for up to an hour, whilst combatting strong winds. As always, it’s easy to make claims. Seeing is believing.

cartoon-nov-2015To answer the question, the biggest challenge is power. Electric propulsion is highly desirable, but battery technology rarely allows beyond 25 minutes practical flight times on multi-rotor and 90 minute flight times on fixed wing. This is the reason we’ve focused heavily on developing fuel cell propulsion systems for fixed wing aircraft. We believe our survey aircraft will eventually be capable of electric flight endurances of over 22 hours.”

Duran de Villiers, director/owner, SteadiDrone said that the challenge is to develop an aerial platform that is able to lift a wide variety of payloads, is compact enough to transport but large enough for higher lift capacity and longer endurance, while at the same time having the system, be easy to fly and operate for the end user. “Our key focus is on development of a range of drones, small enough once folded down with a rapid deploy airframe design, has lifting capacities of up to 25 kg, and endurance of up to an hour. The challenge is to have all these features in one unit. There are many drones good at one, or two of these, but very few that offer all three.”

Hannes van der Walt, CEO, The Gadget Shop: ”It has to be size, flight time and cost. If we look at size, I believe a universal design that will appeal to most customers and that is adaptable to most applications, has to be portable and should be able to take a payload of between 1 to 10 kg. Flight time that is adaptable to most applications needs to be between 20 to 40 minutes, I would say. Lastly, cost for such a drone can be in the range of between R15,000 to R50,000.”

Rolf Schlub, director, TerraCam. “The primary challenge continues to be efficiency of the power source. Although battery technology is advancing, the weight to performance ratio remains a limiting factor for size and endurance. Gasoline powered systems offer endurance but impose weight penalties. Fortunately sensors are decreasing in size and weight and airframes are becoming increasingly versatile and adaptable for both vertical and horizontal flight.”

Peter Cosyn, site manager/directorof R&D, Gatewing (a Trimble company): ”Universal platforms are similar to the universal car: it doesn’t exist and it will likely never exist. The main reason is the payload: size, weight and power requirements are different for different applications. A drone carrying Amazon books will have different requirements than a drone carrying a laser scanner or a drone for photogrammetry. You could envision a very large drone doing everything but it lacks flexibility and the cost will be too high for most applications to have a business case.”

With many flight restrictions in place are there enough opportunities to carry out test flights before large scale manufacture can take place? Are there special licensing arrangements in place for developers and manufacturers? If not, what would you recommend?

Duran de Villiers says that the new drone regulations imposed by the South African Civil Aviation Authority (SACAA) have made it very difficult for a manufacturer to do full test flights. “We have an indoor facility where most of the testing is done for general manufacturing and quality control, but we are forced to travel outside of South Africa to do more in-depth testing. We also work with resellers and clients on testing when possible. There are no ‘arrangements’ or special permission available from the SACAA to date – this may force many larger drone companies and manufacturers to eventually move their business outside South Africa.”

Mark van Wyk ”There are many options.The simplest option is to fly indoors.The SACAA does not regulate the use of remotely piloted aircraft (RPA) indoors. Many international companies are doing significant testing indoors, in large warehouses.

“Alternatively, you may fly a light, remotely piloted aircraft under private use. This means a light RPA (less than 7 kg) no further than 500 m from the pilot, and no higher than the highest obstacle within 300 m). It’s strictly for non-commercial use, and there are other regulations that must be abided by. For more complex operations and testing, operators may contact Denel for testing at the Overberg test base. Unfortunately, there are fees for this that may be prohibitive for smaller operators.

“UAV Industries are negotiating an airspace prohibited from manned aircraft for RPA flight testing. Members will get access to this area, and I understand the costs will be very affordable. They also allow operators to bypass much the hefty compliance process by falling in under their umbrella.

“Lastly, if you plan to operate and wish to have your own operator’s certificate, it may be worthwhile going through the process. You will be able to register your prototype aircraft with the SACAA. You’ll need to provide a safety case, stating that your prototype will only be flown in a manner that provides very little risk of harming animals, people or property or colliding with other aircraft.”
Rolf Schlub: “One can argue that test flights are not directly compensated and hence don’t fall under commercial operations. This might mean that preliminary test flights could be conducted under the South African Model Aircraft Association (SAMAA) regulations. More advanced testing would need consent from the SACAA that should establish dedicated RPAS testing sites similar to that the US has done.”

Peter Cosyn: ”Even in very restricted countries such as Belgium (where the regulation is still pending), there are opportunities to fly on specific test terrains. In general this is sufficient to ensure that a design is tested and risks are captured. The main issue with restrictions is tuning a design for specific market application or user need. Doing specific aerial work tests above relevant terrain or objects, which can be evaluated prior to release of product, is a different story and in most cases not difficult to achieve with a vehicle that is still in a design stage. Specific provisions in national laws that allow such tests (within certain restrictions) would be a step forward.”

There is much talk about using drones for delivering small parcels, even fast food. Do you consider it practical? Would the design of the drone have to be changed substantially from the current ones on the market?

Duran de Villiers: ”Currently delivering anything with a drone accurately, safely and practically is purely marketing fluff, used by companies to create viral marketing campaigns. I get many questions and suggestions for delivery drones, but the technology simply is not there, yet. The very first delivery of any drone was pizza and beer at the Oppi-koppi festival, and was done with a SteadiDrone, the very recent mail delivery in Singapore, a world’s first, was also with a SteadiDrone system. The actual drone design would not need to change much, more the payload bays and systems, safety aspects would need to be addressed obviously, which may lead to some airframe changes, but the real challenge is the navigational and obstacle avoidance technology.”

Mark van Wyk: ”I personally think it’s an excellent idea. This said, I can’t see it happening soon. According to the regulations set by the SACAA, the aircraft may not be flown autonomously. This means it needs to be under constant supervision by a pilot. This requires a data link, which will be hard to maintain in urban areas, especially over long distances. You’ll need special approval from the director of the SACAA to fly near people, over roads, and also to release anything from the aircraft. I believe these are mostly technical problems. When these problems are solved, and an operator can successfully demonstrate to the regulator that he can successfully pilot the aircraft to a drop-off-point, maintaining constant communication, and posting absolutely no thread to animals, people, and property, I’m confident the SACAA will approve the operation. As far as a releasing a payload goes, I think current designs will be easily modified.”

Hannes van der Walt: ”It would be amazing but Google said: “We expect it will take years to develop a service with multiple vehicles flying multiple deliveries per day.” There are obviously many challenges to this – safety being one of them. Where will the drone “drop” the load – on your grass, balcony, or porch? What happens if the drone runs out of battery power during a journey? The design and features would definitely have to change quite substantially. For one, the rotor blades will have to be designed in such a way that it is “safe” to the consumer. These drones will have to be able to detect and avoid obstacles like buildings, trees, people, animals, water, etc. That being said, self-flying vehicles could open up entirely new approaches to moving goods – including options that are cheaper, faster, less wasteful and more environmentally sensitive than what’s possible today.”

Peter Cosyn: ”The business case will only work if drones are highly automated and fly below line of sight (BLOS) without supervision. This would require technology that allows BLOS flights to be performed safely, e.g. by using novel airspace separation methods (NASA UTM project) and/or collision avoidance mechanisms. Drones for parcels will typically be heavier that the typical drone for mapping or surveillance applications so the risks are higher. BLOS concepts should probably be tested with smaller drones first. There are no design changes for drones although we can envision that energy consumption will be even more important for the ‘parcel drone’, driving designs in aerodynamics (for fixed wing version) and power and energy systems.”

Rolf Schlub: ”Intuitive engineering and technology can overcome most of the technical challenges associated with parcel deliveries. Risks and hazards presented by technical malfunctions can largely be mitigated through system redundancy, system awareness and automatic fail-safe procedures. Diverse flying platforms would be required depending on parcel type and delivery location. The foremost challenge, however remain political and social which are far more difficult to mitigate.”

There many options to power a drone: from batteries to fuel cells to hydrogen. In your opinion, what is will be the best option?

All panellist mentioned that lithium type batteries are currently the best option. Here are some individual comments.

Mark van Wyk: “For fixed wing intelligence surveillance and reconnaissance (ISR) missions I believe internal combustion engines are the most reliable and offer long endurance. For shorter missions, batteries are the simplest, cleanest and easiest. The Panasonic NCR18650A is a phenomenal battery and is often used to break flight endurance records. My personal passion is in hydrogen fuel cells. I believe they’re the better option for long endurance flights where noise, vibration or pollution are issues. We’re working on a design for our Alpha aircraft that will allow over 22 hours electric endurance.”

Rolf Schlub: “There will always be a requirement for different types of power delivery systems. These depend on costs, type of mission, noiseand emission restrictions. For long endurance missions hydrogen fuel-cells present the best all round solution while optimised tuned EFI motors also present an attractive alternative. With advancements in solar technology, batteries powered systems will still have their place and appropriateness.”

Duran de Villiers: “Right now batteries are still the most reliable and available. Battery technology is also evolving and will over time offer more energy than current lithium ion and polymer batteries. Fuel cells are of course another option, but again, still needs a lot of development to be practical. ”

As powering by battery appears to be a popular option, what type of battery do you consider the best option? Have you or are you considering a fast-charge option are part of the design?

There was basic agreement on lithium type batteries as the best suited. Here are some more detailed comments:

Peter Cosyn: “Lithium-based batteries are the logical choice because of their high energy density and adequate power density. Today, lithium polymer batteries are the preferred choice but there might be other lithium-based batteries that power the drones of the future. As the battery is an easy replaceable component, fast-charge doesn’t add as much value as it does in a car but it is definitely a feature that will be employed if it can be done without significant reduction in service life of the battery.”

Hannes van der Walt: “Lithium Ion (Li ion) is the fastest growing battery system and is used where high-energy density and lightweight is of prime importance. I would however tread carefully when considering a “fast-charge” option. Although a great selling point, fast charging systems normally have a very negative effect on the overall quality and longevity of most type of batteries. This is due to the fact that fast charges raises the voltage at which a battery is charged at and this results in an internal temperature rise inside of the battery, which is not recommended if you’re hoping to get the most out of your battery.”

Is there a standard for collision avoidance built into every drone or does each manufacturer have its own flavour?

Mark van Wyk: “At the moment it’s a big problem, and in my opinion, yet to be properly solved. This is the main reason the regulator requires a pilot to be present and be in constant control. This is also the reason they’ve restricted drone flying to visual line of sight (VLOS) of 500 m from the pilot and a 400ft above ground. In order to escape this box, operators need to prove to the SACAA that their aircraft can avoid other conflicting aircraft. I’ve heard various claims from developers in South Africa stating they’ve solved the problem, but I haven’t as yet seen a demonstration or heard a good explanation. We’ve been forced to look at developing our own solution, based on a combination of Signals Intelligence (SIGINT) – finding the direction of radio signals from air band radio, transponders, ADS-B transmitters, etc and RADAR. We’re working with local RADAR experts to try and solve this problem.”

Duran de Villiers: “Collision avoidance is where it’s all happening at the moment, the ability for these amazing robots to self-navigate is what’s missing in many of the applications and uses, like delivery. Currently every manufacturer is either using a 3rd party basic system like a standard sonar sensor or developing their own, more advanced systems. There will soon be a few stand-out products available across the market for avoidance sensors, but right now it’s still a bit of a DIY thing for most drones.”

Peter Cosyn: “Today there is no standard and it is unlikely that there will be a standard in the foreseeable future. The reason is that collision avoidance capability scales with size and number of the sensors implemented in the systems. While a standard might be possible for bigger UAS (e.g +25 kg), it will never be a fit for the smaller UAS. Collision avoidance is also a feature that is too much seen as ‘the holy grail’ for avoiding collisions, while in reality it will only be a technology that can reduce collision risk but not avoid it. The speed difference between a fast moving manned aircraft (e.g. fighter jet) and drone is too big to give the drone any chance to out-manoeuvre the manned aircraft. This why making sure that the drone and the aircraft don’t occupy the same part of the airspace (e.g. the NASA UTM project) is more important than a collision avoidance system.”

Rolf Schlub: “There are currently major developments in sense-and-avoid technologies in lieu of integrating drones into commercial airspace. ADS-B transponders also found in manned aircraft will enable detailed tracking of drones by air traffic control. On-board sensors such as sonar, laser rangefinders and precise GPS modules will provide complete situational awareness and give the drone the ability to take reactive measures to prevent collision in the air or with ground obstacles. The regulatory standards defining sense-and-avoid are still several years away.”

Hannes van der Walt: “Most consumer drones are unable to autonomously avoid obstacles. Much research has been devoted to finding a viable solution, but experts say at best the technology is still years away from implementation. This technology will however completely change the flying experience and make drone-based servicesmuch safer.”

If you were the regulator of air space would you allow hobby drones to fly without restrictions? Would you include a minimum training/certification requirement in the regulations?

Mark van Wyk: ”I personally think the SACAA has done a fantastic job of regulating drones. They’ve taken the existing airspace system, which they know and understand – and has proven it’s effectivity over a hundred years of aviation – and adapted it for the operation of unmanned aircraft. I believe that once drones have proven themselves safe, the SACAA will remove some of the constraints and make it easier for new operators to come onto the market. Their approach has to be completely risk averse. I would say a hobby drone would be classified as a toy, for non-commercial use.

“Toys are not regulated by the SACAA. Radio controlled model aircraft can be flown non-commercially with a SAMAA license at SAMAA registered fields, and drone owners are welcome to fly their aircraft under “private use”, non-commercially under RVLOS conditions.I think this leaves lots of room for flying “hobby drones”, and I’d probably recommend the same regulations.

Hannes van der Walt: “I would definitely not allow hobby drones to fly without restrictions. I believe our local laws with regards to drones (both hobby and commercial), as set out by the SACAA, is a pretty good framework and ensures that the safety of everyone is put first and is of utmost priority. Hobby pilots of drones will not need a licence. They will however have to respect limitations imposed on them. The new drone law basically states that hobby pilots:

  • Do not need a remote pilot’s licence.
  • Do not need to register their RPAS aircraft.
  • Do not need Ops Manual, ROC, RLA, etc.
  • Do not need air services licence.
  • Can fly RPAS up to 7 kg weight.
  • Can fly at night (previous not allowed for any RC aircraft).
  • Can fly up to the height of the highest object in 300 m of the drone (or up to a max of 122 m, whichever is the lower of the two).
  • Can fly up to 500 m away from the pilot, while maintaining direct line of sight (RVLOS).
  • Need to keep more than 50 m from people.
  • Need to keep more than 50 m from roads.
  • Need to keep more than 50 m from buildings.
  • Need to keep more than 10 km from an aerodrome (airport or airstrip).
  • May not fly in No Fly Zones.
  • Do not need a handheld fire extinguisher.
  • Do not need a first aid kit.
  • Commercial users of drones will need to license the craft and the pilot.”

Rolf Schlub: “Model aeroplanes, helicopters and other craft have been around for decades and governed under their own regulatory bodies that were meant for hobbyists only. The advent of intelligent flight controllers with GPS that enabled anyone to fly in a way that would normally have taken years of practice is what started to bring the drone movement together.
“No matter what platform hobbyists use and provided they stay within the limits of their rules, I think that should suffice. The problem is that the line between hobbyist and operator is becoming very blurred.”

Peter Cosyn: “It doesn’t make sense to allow hobby drones to fly unrestricted and require restrictive qualifications from the other drone users. The hobby drone community will dwarf the professional drone community and poses the highest thread to our safety. Training/certification needs to be required for the bigger hobby drone while the small ones should comply at least to certain (airspace) restrictions and subscribe at least to certain ‘general rules of good conduct’ that make them aware of the risk of flying in the national airspace. A registration might be needed to enforce responsibility in case of misconduct.”

Duran de Villiers:  “ I would limit the use of hobby drones far more than the use for commercial drones, flown by trained and experienced operators. It’s not the professionals causing the problems, it’s the hobbyist who buys a DJI phantom from a hobby shop and then goes and flies at airports, above crowds etc. I agree that drone operators need licences to own and fly these machines, just like full scale aviation, but the current regulations make it near impossible for the operator, who has been flying responsibly for years, , to earn a living. Many companies have had to shut their doors because of this. The rules needs to be fair and practical.”

Some additional comments from panellists:

Hannes van der Walt: “From a safety point of view, all countries are looking at separating drones from real aircraft, to keep the skies safe for people in real aircraft. South Africa is one of the first countries to regulate the use of commercial drones. DJI is a leading edge company that has built in functionality to keep drone flying safe. They allow the user to set a maximum height and distance for the drone, and they have exclusion areas programmed all over the world, to prevent DJI drones being flown near larger airports. Many of the South African airports are already programmed into each and every DJI drone.”

Mark van Wyk: “I’d like to hear from anyone working on the sense-and-avoid (collision avoidance) problem, and would be more than happy to educate anyone around hydrogen fuel cell propulsion systems. My contact details are on our web site at Also important to note is that the Technology Innovation Agency recognises the importance of aviation in our society. They’re a government grant organisation that gives grants to innovators that are developing technology related intellectual property in our country, doing novel research, resulting in the betterment of South Africa.”

Rolf Schlub: “The use of hobby drones has become widespread and uncontrollable. My recommendation to the SACAA back in July 2012 after meetings with their RPAS team was to start by building a database of users/owners similar to when buying a vehicle, firearm or television. A registration or licence of some type issued at point-of-sale would have allowed the SACAA to monitor sales, usage and growth for the past three years and better develop means to regulate and control them.

“Three years later and due to industry pressure, the SACAA developed legislation far too strict to police and costly to obtain and enforce.

“The legislation which is similar to manned aircraft will impose entry barriers for opportunists but at the same time make many avoid getting compliant at all due to the prohibitive costs involved.”

Duran de Villiers: ”While I agree with a lot of what the SACAA has implemented, I also feel that it’s largely due to control and the money aspect of this new industry, there is no practical need for a person who wants to simply take real estate photos at less than 50 m to have an aviation medical, a restricted radio licence, a full RPL licence, aviation exams, yet the hobbyist can fly the same drone, with the same equipment in more or less the same situation without any of the above. The thing is, people have been flying drones in this country for many years, with no issues, I’ve flown on many film shoots, commercials, TV productions, aerial photography and more without a single incident, these regulations, as needed as they may, may also kill the very industry it’s trying to ‘protect’. South Africa has some great companies in this new industry, doing great things internationally, but we cannot even test fly in our own country. SteadiDrone was ‘Proudly South African’ registered but how can we be if we need to travel abroad to do most of our business…”

Thank you to our panellists. While regulations are necessary, South Africa should not over-regulate which is often the tendency. The drone industry is an interesting development to follow and shows great potential. What is your view about the various aspects of drones? The panellists have shared their views, it is now over to you. Drop us an email at We welcome your comments.

Related Articles

  • South African Government COVID-19 Corona Virus Resource Portal
  • Now Media acquires EngineerIT and Energize from EE Publishers
  • Printed electronics: The defining trends in 2019
  • Charlie and the (fully-automated) Chocolate Factory
  • SANSA app calculates best HF communication channel