One-Motor Drone Mimics Maple Seeds For Stability

We’ve seen aircraft based on “helicopter” seeds (technically samara seeds, which include those of maples and elms) before, but this recent design from researchers at the Singapore University of Technology and Design (SUTD) shows how a single small motor can power a spinning monocopter capable of active directed flight, including hovering.

The monocopter is essentially an optimized wing shape with a single motor and propeller at one end. Hardware-wise it might be simple, but the tradeoff is higher complexity in other areas. Physical layout and balance are critical to performance, and software-wise controlling what is basically a wing spinning itself at high speed is a complex task. The payoff is highly-efficient flight in a package that self-stabilizes; it weighs only 32 grams and has a flight time of 26 minutes, which is very impressive for a self-contained micro aircraft.

We saw what looks like an earlier version of this concept from SUTD that was capable of directed flight by modifying the airfoil surface, but like the seeds it was modeled after, it’s more of a glider. This unit has the same spinning-seed design, but is actively powered. A significant improvement, for sure.

For those who prefer their DIY micro aircraft a little more traditional-looking, be sure to check out the design details of a handmade and fully operational 1:96 scale P-51 Mustang that weighs only 2.9 grams. It even has retractable landing gear! When one can manage to keep mass to a bare minimum, a little power goes a long way.

Student Drone Flies, Submerges

Admit it. You’d get through boring classes in school by daydreaming of cool things you’d like to build. If you were like us, some of them were practical, but some of them were flights of fancy. Did you ever think of an airplane that could dive under the water? We did. So did some students at Aalborg University. The difference is they built theirs. Watch it do its thing in the video below.

As far as we can tell, the drone utilizes variable-pitch props to generate lift in the air and downward thrust in water. In addition to the direction of the thrust, water operations require a lower pitch to minimize drag. We’d be interested in seeing how it is all waterproofed, and we’re unsure how deep the device can go. No word on battery life either. From the video, we aren’t sure how maneuverable it is while submerged, but it does seem to have some control. It wouldn’t be hard to add a lateral thruster to improve underwater operations.

This isn’t the first vehicle of its kind (discounting fictional versions). Researchers at Rutgers created something similar in 2015, and we’ve seen other demonstrations, but this is still very well done, especially for a student project.

We did see a submersible drone built using parts from a flying drone. Cool, but not quite the same.

Continue reading “Student Drone Flies, Submerges”

A screenshot of the software in action is shown. A sidebar on the left shows an icon of a skull-shaped drone above the text “DAMN VULNERABLE DRONE.” Below this, it lists controls for the simulator, and resources for using the software. In the rest of the screen, a rendered scene is shown. A rendered computer monitor showing “DRONE HACKER” is at the bottom of the scene. Above this is a hovering drone, and behind it is a table labeled “Ground Control Station” with a man sitting at it.

A Vulnerable Simulator For Drone Penetration Testing

The old saying that the best way to learn is by doing holds as true for penetration testing as for anything else, which is why intentionally vulnerable systems like the Damn Vulnerable Web Application are so useful. Until now, however, there hasn’t been a practice system for penetration testing with drones.

The Damn Vulnerable Drone (DVD, a slightly confusing acronym) simulates a drone which flies in a virtual environment under the command of of an Ardupilot flight controller. A companion computer on the drone gives directions to the flight controller and communicates with a simulated ground station over its own WiFi network using the Mavlink protocol. The companion computer, in addition to running WiFi, also streams video to the ground station, sends telemetry information, and manages autonomous navigation, all of which means that the penetration tester has a broad yet realistic attack surface.

The Damn Vulnerable Drone uses Docker for virtualization. The drone’s virtual environment relies on the Gazebo robotics simulation software, which provides a full 3D environment complete with a physics engine, but does make the system requirements fairly hefty. The system can simulate a full flight routine, from motor startup through a full flight, all the way to post-flight data analysis. The video below shows one such flight, without any interference by an attacker. The DVD currently provides 39 different hacking exercises categorized by type, from reconnaissance to firmware attacks. Each exercise has a detailed guide and walk-through available (hidden by default, so as not to spoil the challenge).

This seems to be the first educational tool for drone hacking we’ve seen, but we have seen several vulnerabilities found in drones. Of course, it goes both ways, and we’ve also seen drones used as flying security attack platforms.

Continue reading “A Vulnerable Simulator For Drone Penetration Testing”

Picture of self landing drone satellite with orange and black body. Propellors are extended.

FPV Drone Takes Off From A Rocketing Start

Launching rockets into the sky can be a thrill, but why not make the fall just as interesting? That is exactly what [I Build Stuff] thought when attempting to build a self-landing payload. The idea is to release a can sized “satellite” from a rocket at an altitude upwards of 1 km, which will then fly back down to the launch point.

The device itself is a first-person view (FPV) drone running the popular Betaflight firmware. With arms that swing out with some of the smallest brushless motors you’ve ever seen (albeit not the smallest motor), the satellite is surprisingly capable. Unfortunately due to concerns over the legality of an autonomous payload, the drone is human controlled on the descent.

Using collaborated efforts, a successful launch was flown with the satellite making it to the ground unharmed, at least for the most part. While the device did show capabilities of being able to fly back, human error led to a manual recovery. Of course, this is far from the only rocketry hack we have seen here at Hackaday. If you are more into making the flight itself interesting, here is a record breaking one from USC students.

Continue reading “FPV Drone Takes Off From A Rocketing Start”

Triggering Lightning And Safely Guiding It Using A Drone

Every year lightning strikes cause a lot of damage — with the high-voltage discharges being a major risk to buildings, infrastructure, and the continued existence of squishy bags of mostly salty water. While some ways exist to reduce their impact such as lightning rods, these passive systems can only be deployed in select locations and cannot prevent the build-up of the charge that leads up to the plasma discharge event. But the drone-based system recently tested by Japan’s NTT, the world’s fourth largest telecommunications company, could provide a more proactive solution.

The idea is pretty simple: fly a drone that is protected by a specially designed metal cage close to a thundercloud with a conductive tether leading back to the ground. By providing a very short path to ground, the built-up charge in said cloud will readily discharge into this cage and from there back to the ground.

To test this idea, NTT researchers took commercial drones fitted with such a protective cage and exposed them to artificial lightning. The drones turned out to be fine up to 150 kA which is five times more than natural lightning. Afterwards the full system was tested with a real thunderstorm, during which the drone took a hit and kept flying, although the protective cage partially melted.

Expanding on this experiment, NTT imagines that a system like this could protect cities and sensitive areas, and possibly even use and store the thus captured energy rather than just leading it to ground. While this latter idea would need some seriously effective charging technologies, the idea of proactively discharging thunderclouds is perhaps not so crazy. We would need to see someone run the numbers on the potential effectiveness, of course, but we are all in favor of (safe) lightning experiments like this.

If you’re wondering why channeling lightning away from critical infrastructure is such a big deal, you may want to read up on Apollo 12.

Improving Flying Drones By Mimicking Flying Squirrels

With the ability to independently adjust the thrust of each of their four motors, quadcopters are exceptionally agile compared to more traditional aircraft. But in an effort to create an even more maneuverable drone platform, a group of South Korean researchers have studied adding flying squirrel tech to quadcopters. Combined with machine learning, this is said to significantly increase the prototype’s agility in an obstacle course.

Flying squirrels (tribe Pteromyini)) have large skin flaps (patagium) between their wrists and ankles which they use to control their flight when they glide from tree to tree, along with their fluffy squirrel tail. With flights covering up to 90 meters, they also manage to use said tail and patagium to air brake, which prevents them from smacking with bone jarring velocities into a tree trunk.

By taking these principles and adding a similar mechanism to a quadcopter for extending a patagium-like membrane between its rotors, the researchers could develop a new controller (thrust-wing coordination control, TWCC), which manages the extending of the membranes in coordination with thrust from the brushless motors. Rather than relying on trial-and-error to develop the controller algorithms, the researchers trained a recurrent neural network (RNN) which was pre-trained prior to first flights using simulation data followed by supervised learning to refine the model.

During experiments with obstacle avoidance on a test-track, the RNN-based controller worked quite well compared to a regular quadcopter. A disadvantage is of course that the range of these flying squirrel drones is less due to the extra weight and drag, but if one were to make flying drones that will perch on surfaces between dizzying feats of agility in the air, this type of drone tech might just be the ticket.

Continue reading “Improving Flying Drones By Mimicking Flying Squirrels”

Supercon 2024: Killing Mosquitoes With Freaking Drones, And Sonar

Suppose that you want to get rid of a whole lot of mosquitoes with a quadcopter drone by chopping them up in the rotor blades. If you had really good eyesight and pretty amazing piloting skills, you could maybe fly the drone yourself, but honestly this looks like it should be automated. [Alex Toussaint] took us on a tour of how far he has gotten toward that goal in his amazingly broad-ranging 2024 Superconference talk. (Embedded below.)

The end result is an amazing 380-element phased sonar array that allows him to detect the location of mosquitoes in mid-air, identifying them by their particular micro-doppler return signature. It’s an amazing gadget called LeSonar2, that he has open-sourced, and that doubtless has many other applications at the tweak of an algorithm.

Rolling back in time a little bit, the talk starts off with [Alex]’s thoughts about self-guiding drones in general. For obstacle avoidance, you might think of using a camera, but they can be heavy and require a lot of expensive computation. [Alex] favored ultrasonic range finding. But then an array of ultrasonic range finders could locate smaller objects and more precisely than the single ranger that you probably have in mind. This got [Alex] into beamforming and he built an early prototype, which we’ve actually covered in the past. If you’re into this sort of thing, the talk contains a very nice description of the necessary DSP.

[Alex]’s big breakthrough, though, came with shrinking down the ultrasonic receivers. The angular resolution that you can resolve with a beam-forming array is limited by the distance between the microphone elements, and traditional ultrasonic devices like we use in cars are kinda bulky. So here comes a hack: the TDK T3902 MEMS microphones work just fine up into the ultrasound range, even though they’re designed for human hearing. Combining 380 of these in a very tightly packed array, and pushing all of their parallel data into an FPGA for computation, lead to the LeSonar2. Bigger transducers put out ultrasound pulses, the FPGA does some very intense filtering and combining of the output of each microphone, and the resulting 3D range data is sent out over USB.

After a marvelous demo of the device, we get to the end-game application: finding and identifying mosquitoes in mid-air. If you don’t want to kill flies, wasps, bees, or other useful pollinators while eradicating the tiny little bloodsuckers that are the drone’s target, you need to be able to not only locate bugs, but discriminate mosquitoes from the others.

For this, he uses the micro-doppler signatures that the different wing beats of the various insects put out. Wasps have a very wide-band doppler echo – their relatively long and thin wings are moving slower at the roots than at the tips. Flies, on the other hand, have stubbier wings, and emit a tighter echo signal. The mosquito signal is even tighter.

If you told us that you could use sonar to detect mosquitoes at a distance of a few meters, much less locate them and differentiate them from their other insect brethren, we would have thought that it was impossible. But [Alex] and his team are building these devices, and you can even build one yourself if you want. So watch the talk, learn about phased arrays, and start daydreaming about what you would use something like this for.

Continue reading “Supercon 2024: Killing Mosquitoes With Freaking Drones, And Sonar”