The last few days have seen drone stories in the news, as London’s Gatwick airport remained closed for a couple of days amid a spate of drone reports. The police remained baffled, arrested a couple who turned out to be blameless, and finally admitted that there was a possibility the drone could not have existed at all. It emerged that a problem with the investigation lay in there being no means to detect a drone beyond the eyesight of people on the ground, and as we have explored in these pages already, eyewitness reports are not always trustworthy.
Radar Can’t See Them
It seems odd at first sight, that a 21st century airport lacks the ability to spot a drone in the air above it, but a few calls to friends of Hackaday in the business made it clear that drones are extremely difficult to spot using the radar systems on a typical airport. A system designed to track huge metal airliners at significant altitude is not suitable for watching tiny mostly-plastic machines viewed side-on at the low altitudes. We’re told at best an intermittent trace appears, but for the majority of drones there is simply no trace on a radar screen.
We’re sure that some large players in the world of defence radar are queueing up to offer multi-million-dollar systems to airports worldwide, panicked into big spending by the Gatwick story, but idle hackerspace chat on the matter makes us ask the question: Just how difficult would it be to detect a drone in flight over an airport? A quick Google search reveals multiple products purporting to be drone detectors, but wouldn’t airports such as Gatwick already be using them if they were any good? The Hackaday readership never fail to impress us with their ingenuity, so how would you do it?
Can You Hear What You Can’t See?
It’s easy to pose that question as a Hackaday scribe, so to get the ball rolling here’s my first thought on how I’d do it. I always hear a multirotor and look up to see it, so I’d take the approach of listening for the distinctive sound of multirotor propellers. Could the auditory signature of high-RPM brushless motors be detected amidst the roar of sound near airports?
I’m imagining a network of Rasberry Pi boards each with a microphone attached, doing some real-time audio spectrum analysis to spot the likely frequency signature of the drone. Of course it’s easy to just say that as a hardware person with a background in the publishing business, so would a software specialist take that tack too? Or would you go for a radar approach, or perhaps even an infra-red one? Could you sense the heat signature of a multirotor, as their parts become quite hot in flight?
Whatever you think might work as a drone detection system, give it a spin in the comments. We’d suggest that people have the confidence to build something, and maybe even enter it in the Hackaday Prize when the time comes around. Come on, what have you got to lose!
The link uses standard WiFi hardware in a slightly unusual way to create a digital data link that acts more like an analog system, with a preference for delivering low latency video and a graceful drop-off when signal quality gets poor. A Raspberry Pi Zero, Alfa NEH WiFi card, external antenna, battery, and a 3D printed enclosure result in a self-contained unit. Two are needed: one for each end of the link. One unit goes on the drone and interfaces to the flight controller, and the other is for the ground station.
A companion android app allows for just about any old Android phone to serve as video feed, on-screen display of telemetry data, and touchscreen interface.
The software is DroneBridge (GitHub repository) and it implements Wifibroadcast which uses WiFi radios, but without the usual WiFi functionality. A Raspberry Pi is the usual platform, but there’s also an ESP32 port. The software is capable of even more, but so far suits [GlytchTech]’s needs just fine, and he was able to refine his original Watch_Dogs-inspired hacking drone with it.
Picture this: it’s late in the evening on a freezing cold, dark, and windy December night in southern England, and an airport worker at Gatwick — London’s second international airport — sees something fly past in the gloom above the floodlights. The weather and darkness makes it difficult to see what the object was, but the report is phoned in to security. What was it? A flock of birds? A piece of plastic litter caught by the wind and blown through the night? In this case, the call is recorded as a drone. Because the magic D-word has been uttered, a security plan swings into action, the airport is put on a high state of readiness, and flights are suspended.
Thousands of people across the site are put on alert, watching for the drone. And of course, the drone reports roll in, and the story takes on a life of its own. People who have no idea what a drone looks like in the air are now expecting to see one, so of course when a flock of birds or a plastic bag caught by the wind crosses their peripheral vision they too are convinced that it is the drone. Night turns into day, there is a lull in the reports so the airport re-opens, only to be closed again following a fresh spate of sightings. Flights are diverted all across the country, and tens of thousands of passengers are stranded in the terminals.
There follows three days of airport closure drama. No photos emerge despite almost every one of the many thousands of people on the site having a camera phone from which they are Tweeting about the queues in the terminal. There is a grainy video, but it is indistinct, and crucially it doesn’t have anything in it that is identifiable as Gatwick. Meanwhile the police are frustrated in their search for the drone operators, who like their drone, prove difficult to pinpoint
You might imagine that this was the fictional plot of a thriller novel, but sadly not. All of the above is a tale of the last few days of events in the British news, save for most of the first paragraph which is our guess at how the first drone sightings may have happened. At the time of writing there remains the possibility that there could have been a drone over Gatwick, but given the current dearth of evidence it is one that seems tenuous. There are reports of drone wreckage, but since readers with long memories will recall UK police once identified RepRap parts as a 3D printed gun we’ll wait until we see it before we call it that.
If there was indeed a drone then of course we would like to see its operators brought to justice forthwith. But what concerns us at Hackaday are the implications the episode could still have for those of our community with an interest in multirotors. The usual clamour was made for Government to do something about it, and we know that would have meant a fresh set of onerous regulations for responsible multirotor owners while doing nothing about the criminals, because of course criminals have little regard for laws.
So if we are to glean anything from this sorry mess, we must examine it from several angles. Why is there a lack of drone detection technology in place? How should drone reports initially be treated and investigated on the ground? How should they be dealt with in official inquiries, and how then should lawmakers see them? This will inevitably have a British flavour to it because of the incident in question, but the points are just as valid worldwide.
When a Drone Report Comes In, We Need a Reliable Way to Evaluate It
When we are told something new, it passes a process of evaluation in our minds. We look at the source, and weigh up the story itself. If a guy with crazy hair in the street tells us that the aliens have landed and are controlling the Prime Minister with a ray gun, it will probably be discounted. But if Hackaday tells us that someone has hacked a VGA chipset to work as a software-defined radio we’re guessing most of you would be very interested indeed.
When a fresh drone incident is reported it appears that this evaluation process has historically been defective. We have previously discussed official incident reports that come with no physical evidence of a drone, but contain descriptions of drones with capabilities unmatched even by jet fighter aircraft. It seems like any eyewitness report in which the culprit is named as a drone is automatically taken at face value no matter how unlikely it may be. The fact that a report may have come from a pilot is sometimes mentioned as a boost to its credibility, but that is a false assumption. A pilot who is not familiar with either how drones appear from a distance or what the capabilities of a drone are in the air can only be considered an unreliable witness, because while they may know a lot about aircraft they lack the required expertise for this judgement. So what can be done to help boost the quality of reporting and to immediately highlight credible reports while requiring more for dubious ones?
In the case of a near miss in open airspace there may be little effect on ground-based facilities, but at an airport such as Gatwick there can be no chances taken by the authorities. A drone collision on an aircraft on final approach could cause hundreds of fatalities, so upon receipt of a report they must have had little choice but to close the runways. There appears to have been a lack of drone detection technologies in place at Gatwick which means that the only source available to the airport would have been the eyewitnesses themselves, and since we have amply demonstrated the potential for eyewitness reports being unreliable then the current confusion becomes an inevitability. It is imperative that more reliable detection technologies be fitted or developed if necessary. This is especially true when precautionary shut-downs stretch past minutes or hours into world-news-making delays as happened in this instance.
Competent Police Investigations and Responsible Journalism on Drone Reports
Once an incident has started and news of it emerges there is a consequent effect upon members of our community. Legitimate drone fliers away from the airport will find themselves under more scrutiny, and since it is already a common tale to hear of police being called when flying is under way that means they could face harassment and wrongful arrest. Indeed though we do not know all the details of the pair arrested near Gatwick it smacks of their being arrested in a round-up of convenient local drone enthusiasts rather than as a result of meaningful investigation. That the names of the pair were leaked and they became the subject of a media frenzy further shows the danger in which they were placed, as well as the irresponsibility of the reporters who covered their plight.
Perhaps Most Importantly: We Need Accurate Official Incident Reports
Whatever happens in a drone report, whether it be an arrest or an embarrassing debacle, there will inevitably be an official incident report from the Civil Aviation Authority, the regulator of British civilian airspace. This will form the official record of the event, and thus should strive to be as accurate as possible, but here the process falls short for the final time. There appears to be no evaluation step performed on the available evidence and no requirement for physical proof. So if an eyewitness reports behaviors about the drone that no drone ever built could possibly be capable of, it is solemnly recorded as fact. Our previous article on this subject highlights multiple such accounts, and this is an important point because as the official record these reports are what informs legislators. When they make laws pertaining to drones it is imperative that their decisions are based upon accurate evidence, and it is clear that this is not the case. Given that they will no doubt be reviewing drone legislation in the wake of this fiasco it is particularly important that the investigators consult people with specialist knowledge in the field, demand physical proof rather than heresay, and most importantly question accounts that stretch credibility.
It seems obvious that the multirotor hobbyist is caught in a perfect storm of incompetent authorities, deeply flawed investigations, shoddy journalism, and clueless legislators. This incident has laid bare some of the shortcomings, and it is to be hoped that a few lessons might be learned to produce less of a debacle surrounding future drone incidents. It is still a developing story so there may be a breakthrough and the whole narrative will change, and if that turns out to be the case then we hope they find the correct perpetrator this time and send them away at Her Majesty’s pleasure for a very long time. We’re guessing though that every effort will be made to push it as far under the carpet as possible to save red faces among officialdom. As multirotor enthusiasts we must keep the issue of poor investigation alive though, for if we let it be buried once more it will come back to trouble us again.
We love a good drone build here at Hackaday, but no matter how much care is taken, exposed propellers are always a risk: you don’t have to look far on the web to see videos to prove it. Conventional prop-guards like those seen on consumer drones often only protect the side of the propeller, not the top, and the same problem goes for EDFs. [Stefano Rivellini]’s solution was to take some EDFs and place them in the middle of large carbon fibre thrust tubes, making it impossible to get anywhere near the moving parts. The creation is described as a bladeless drone, but it’s not: they’re just well hidden inside the carbon fibre.
We’re impressed by the fact that custom moulds were made for every part of the body, allowing [Stefano] to manually create the required shapes out of carbon fibre cloth and epoxy. He even went to the trouble of running CFD on the design before manufacture, to ensure that there would be adequate thrust. Some DJI electronics provide the brains, and there’s also a parachute deployment tube on the back.
Whilst there’s no doubt that the finished drone succeeds at being safe, the design does come at the cost of efficiency. The power electronics needed are far more serious than we’d usually see on a drone of this size, to compensate for the extra mass of the thrust ducts and the impediment to the air-flow caused by the two 90° bends.
We’ve written on reports of drone near-misses with aircraft here back in 2016, and indeed we’ve even brought news of a previous runway closure at Gatwick. But it seems that this incident is of greater severity, over a much longer period, and even potentially involving more than one machine. The effect that it could have on those in our community who are multirotor fliers could be significant, and thus it is a huge concern aside from the potential for mishap in the skies above London’s second largest airport.
Bees. The punchline to the title is bees carrying sensors like little baby bee backpacks. We would run out of fingers counting the robots which emulate naturally evolved creatures, but we believe there is a lot of merit to pirating natural designs, but researchers at the University of Washington cut out the middle-man and put their sensors right on living creatures. They measured how much a bee could lift, approximately 105 milligrams, then built a sensor array lighter than that. Naturally, batteries are holding back the design, and the rechargeable lithium-ion is more than half of the weight.
When you swap out brushless motors for organics, you gain and lose some things. You lose the real-time control, but you increase the runtime. You lose the noise, but you also lose the speed. You increase the range, but you probably wind up visiting the same field over and over. If your goal is to monitor the conditions of flowering crops, you may be ready to buy and install, but for the rest of us, dogs are great for carrying electronics. Oh yes. Cats are not so keen. Oh no.
If you’re doing remote controlled flight, odds are you’re also flying FPV. Or you at least have a camera on board. If you’re transmitting to the ground, you may have noticed the antenna on your plane has some weird radiation patterns; bank your plane to the left or right, and your signal gets worse. [Ant0003] over on Thingiverse has a great solution to this problem that’s small, lightweight, and will fit into just about any airframe.
[Ant]’s flying a Mini Talon with FPV, and since planes turn slower than drones, and can fly much further than multicopters, the radiation pattern of the antenna is very important. In this case, [Ant] wants to keep the antenna perpendicular to the ground. This problem was solved with a cheap 9-gram servo and a few 3D printed parts that hold an SMA connector. One end of this wire goes to the video transmitter, and the antenna is screwed into the other end.
A servo alone does not make the antenna point straight up. To do this, [Ant] needed to program his flight controller. He’s using iNav, and a few clicks of the mouse makes one servo channel do whatever the gyroscope isn’t doing. The results (video below) speak for themselves. It’s an antenna that always points straight up, which is exactly what this video transmitter needed.