When we think of robotics, the first thing that comes to mind for many of us is a kind of industrial arm stuck on the floor, or perhaps a semi-autonomous rover crossing the dusty landscape of Mars. Although these two environments may be almost different, the basic “rules” are almost the same. The solid ground gives the robot a clear understanding of its position and orientation, which greatly simplifies tasks such as avoiding collisions or communicating with nearby objects.
But what if that reference point is gone? How does a robot navigate when it flies through open space or hover in mid-air? This is the only issue that fascinates Nick Rehm, who came to organize this week Aerial Robotics Hack Chat Talk about his passion for flying robots. He is currently an aerospace engineer at Johns Hopkins Applied Physics Laboratory, where he works on the unique challenges facing autonomous flying vehicles such as mid-air collision detection and avoidance, as well as vertical take-off and landing (VTOL) development. But before his masters in aerospace engineering and rotorcraft, he started out like many of us by playing with DIY projects.
In fact, regular hackade readers will probably remember seeing some of his impressive builds. Its autonomous ekranoplan, designed to pursue a goal using computer vision, hit the front page in April. In 2020, we looked at its entertainment of SpaceX’s Starship prototype, which used a realistic layout of control surfaces and vector-based thrust to perform the spacecraft’s signature “belly flop” strategy – albeit with RC motors and propellers instead of rocket engines. But even before that, Nick asked his mother for permission to separate a Wii controller so that she could use its inertia unit (IMU) on a wooden framed tricycle.
Talking about creating some of these hobbies leads Chat to Nick’s dRehmFlight project, a GPLv3 licensed flight control package that can run on relatively low-cost hardware, such as a Teensy 4.0 microcontroller with GY-521 MPU6050 IMU. The project is designed so that amateurs can easily experiment with VTOL craft, especially those that move between vertical and horizontal flight profiles, and have driven most of Nick’s own flying craft.
Moving on to more technical questions, Nick says that one of the most difficult aspects when designing an autonomous flying vehicle is to ignore your limitations. What he meant by that was that the craft had a clear goal of what to do and, critically, how long it should take. How far should the craft be able to fly? How fast? Do I have to travel to the target location, and if so, for how long? The answers to these questions will determine the final vehicle form, and the key to determining whether the complexity of the transition from VTOL to a fixed-wing horizontal flight is worth implementing.
But according to Nick, the biggest challenge in aerial robotics is onboard state estimation. That is, the ability to know the position and orientation of the craft relative to the soil. Although high-performance computers have become lighter and sensors have improved, he says there is still no substitute for having a ground-based tracking system. He mentioned that the fancy exhibits you have seen flying in the formation of drones and working collaboratively towards a task will almost certainly have an array of motion capture cameras next to it. This makes for an impressive display, but greatly limits the practical application of this drone fleet.
So what does the future hold for aerial robotics? Nick says open source projects like ArduPilot and PX4 are still great choices for amateurs, but he sees promise in new platforms that combine traditional autopilots with more onboard computing power, such as Auterion’s Skynode. More powerful flight controllers can enable simultaneous localization and mapping (SLAM) techniques, which use 3D scans of the environment to steer the robot towards itself. He is also very interested in technology that enables autonomous flight in GPS-denominated environments, which is important for robotic craftsmen who have to work indoors or in situations where they are unavailable or unreliable on satellite. In light of NASA’s incredible success Cleverness Helicopters, we imagine that these techniques will play an invaluable role in future airborne exploration of Mars.
We would like to thank Nick for hosting this week Aerial Robotics Hack Chat, Which has become one of the fastest hours in recent memory Her experience as an avid hobbyist and professional in the field provides just the kind of insights for the Hackade community and her enthusiastic offer to keep in touch with those who participated in the chat to further discuss their projects. He is about this subject. We hope to see some great things from Nick up front, and we want him to join us again in the future to see what he does.
Hack Chat is a weekly online chat session hosted by leading experts from all corners of the hardware hacking universe. This is a great way for hackers to connect in a fun and informal way, but if you can’t make it live, make sure you don’t miss this overview post as well as the transcripts posted on Hackaday.io.