UAVs - Are you considering using them?
As we all know UAV's are often in the news for a multitude of reasons. More specifically the multi-copter type are becoming very popular amongst hobbyists and increasingly for commercial photographic work. BI intelligence estimates the global Aerial Drone market for civilian use could be worth $4bn by 2024 (Ref1). That’s an impressive 30% of the pure military market. UAV technology is one driver, but also the miniaturization and increase in quality of the video technology is another. With the use of FPGA technologies organizations are increasingly able to realize broadcast quality video with point to point links. For example a UK based organization with its own COFDM based modulation and demodulation technology was behind the impressive video captured by the BBC on the back of an Eagle flying down from the Burj Khalifa sky scraper in Dubai earlier in the year. (Ref2).
It is my view those looking to take on the services of a drone / UAV service provider need to dig down far more deeply than at present and take responsibility for ensuring all that can be done is being done in the name of safety. Not only to protect the general public, but to also protect and grow the commercial UAV market sector by gaining public sector confidence. I believe a look under the bonnet of the increasingly complex machines that are being put up into the skies is necessary; be that with the help of a consultant or by personally learning about the technology. An organization’s reputation and the reputation of the UAV market generally can be greatly damaged if this type of due diligence is not done.
The focus of this discussion is to point out some basic areas to look at particularly for electric multi-copters and electric fixed wing or mini-UAVs as I will call them.
Of course safety very much depends on application. From fixed wing UAVs surveying crops right up to multi-copters covering highly populated sporting events, clearly the safety considerations vary greatly. These are some of the questions I think ought to be asked, but right now very few can be adequately answered – certainly within the commercial UAV space:
System Design
a) Is recovery from one or more motor failures possible?
For a multi-copter with four rotors, no. With 6, or 8 or more there are possibilities, but this must be taken into consideration by the flight controller. For a fixed wing UAV there are options which will be looked at later.
b) Is there redundancy of flight control electronics?
Many multi-copters I’ve seen simply install a DJI flight controller or similar and that’s it. What if it were to go down? In fact there are redundancy methodologies that can be implemented relatively easily. Such as a three controllers architecture with a low complexity (CPLD based i.e. non software solution for better reliability) voting hardware that monitors the outputs and passes on the control signals based on at least two controllers ‘agreeing’.
c) Is there a redundant battery?
If a main power battery should fail is it possible to immediately switch to another? (The technology is there to do it).
d) What level of autonomy does the UAV have?
The number of axes of stabilization the unit has or doesn’t have, ought to be understood.
- No stabilization (needs a very skilled pilot)
- 3 axis: Gyro only (alleviates the pilot skill to some extent, especially in bad weather, ref3)
- 6 axis: Gyro + accelerometer flight controller (hands free flying, i.e. it understands which way is up)
- 7 axis: Includes barometer for altitude setting
- 10 axis: Includes GPS for near exact positional flying - Preferred solution for serious photography (What happens if the GPS signal is lost in this case?)
e) Should you gain peace of mind if the supplier or operator says they have a parachute on a multi-copter in case of engine or battery failure? (There are such solutions out there).
Maybe, provided the following questions are adequately answered:
- What is the minimum height that the chute can be effectively deployed?
- What is the UAV fall rate with the chute deployed? And can that still hurt if it lands on someone?
- Is there a mechanism for auto battery disconnect or some other method to ensure the rotors are not rotating during chute deployment and subsequent decent?
- Of course no chute will mitigate a UAV somehow getting onto a busy motorway.
Build Quality
For both mini-UAV and multi-copter the component quality is important. One of the most stressed pieces of the electronics is the ESC (motor controller) which pulses very high current to the windings of the brushless motor(s). Are these fabricated with known (traceable) electronic components from original silicon manufacturers? A scare in the general aviation industry some years back highlights the issue of using low quality copy or pirated parts. How about the motors themselves, are they highly accurately machined titanium motors or the low cost hobbyist versions?
Due diligence on the electrical system
a) It isn’t just a matter of plugging things in and expecting everything to work. Every system is different. Is the high current pulse motor driver isolated from the sensitive transmitter / receiver and control electronics? If not totally isolated with independent batteries is it at least isolated via opto-electronics or with some other method? How about the grounding scheme? Is it carefully thought through to avoid noise and ground loops?
b) Often standard gauge cables are used for the connection of battery and motors. Are these adequate? Has any analysis been done on actual system power usage? Cables that are too thin will operate at higher temperatures, increasing resistance and draining the battery as energy is lost in heat, or worse starving the motor or servos of needed power.
c) Especially for multi-copters which normally rely on a single battery, what is the battery technology? LiPo, LiFe or other? When many such batteries come to their end of capacity they tend to fall off a cliff. Meaning the voltage drops off all of a sudden, making it difficult to assess when the battery is running low. (Integration of actual current used is much better than simple voltage level or duration checks). Are there any checks in place to ensure the LiPo is not run down too far? Regular depletion greater than around 80% of the total charge could certainly store up problems for the LiPo battery in future. Which is the safest technology? Certainly LiFe batteries have a lot going for them.
d) A mini-UAV should certainly have one battery for the motor and one for the other electronics as minimum. Pointless of course for a multi-copter unless a parachute strategy or some other safety feature is used in case of a main battery or motor failure.
e) For mini-UAV has the current budget been properly calculated? Many servos for flight control surfaces don’t even state current requirements (which anyway also varies with flight conditions).
f) Many low cost transmitter / receiver systems today have a telemetry feature for useful real time flight information such as airspeed and temperature. Is the temperature of the ESC (motor driver) and other electronics being monitored? Is the airflow sufficient to cool them adequately? An over temperature condition can certainly send a craft quickly to the ground.
Supplier Credibility
Check out the credentials / reputation /ethics of individuals and/or organizations you engage with. At Bluelight we once contracted a Thailand based (non Thai) so called ‘expert’ to build a prototype of the mini UAV we designed, only to find out half way through, with the majority of our funds spent, they ‘ran away’ onto some other contract and we got nothing at all for our money.
Another example; at a broadcast trade show recently the sales man didn’t even know what a gyro was and still tried to convince me on the safety of the multi-copter.
Fixed wing mini-UAV or Multi-copter?
Depending on the application careful thought should be given to whether it is a multi-copter that is required or could a mini-UAV do the job.
As mentioned above a mini-UAV will likely have a separate battery for the flight electronics. There is a greater chance of the very high current power systems battery failing. If this does fail sensible things can still be done. For a multi-copter a failed battery means the unit has to come down where it is. A fixed wing aircraft is only better if it is properly designed. Many fixed wing aircraft with a motor failure (and no sensible control actions taken) will stall and enter into a spin and crash. But did you know that even with all control surfaces in a central / default position an aircraft can be designed such that it will naturally glide down? This then opens up the possibility for a smart imaging system to steer the craft down to an unpopulated area such as a field or into trees.
Given a choice of fixed wing or multi-copter, I think by far the safest option is fixed wing, with additional benefits of greater range and flight time. With advances in brushless stabilized gimbal mounts this is certainly worth considering even for fixed location photographic requirements. I’ve seen a few VTOL (Vertical Take Off and Landing) aircraft become available. To some extent this offers the best of both worlds and we may see more of these coming onto the market. Of course there are also other interesting options for specific markets to look out for. Blimps for example (ref4).
Conclusion
It’s my view that safety will play an increasingly important role in the commercial UAV market and much work needs to be done if the plans of organizations such as Apple and Amazon are to become a reality.
References
Ref1: Drone Market: http://uk.businessinsider.com/uav-or-commercial-drone-market-forecast-2015-2?r=US&IR=T
Ref2: BBC flying Eagle video: https://www.youtube.com/watch?v=4uJnivXGy0k
Ref3: Bluelight Gyro test: https://www.youtube.com/watch?v=hW3rEfHM0j8
(BL-3GRC Supplied in US by Xflight Technologies http://xflighttech.com/)
Ref4: Blimp example: https://www.youtube.com/watch?v=2jQSF5IRq3A
Replies