[FlorianH] Shows Off MinimaBL, The Next Generation Of His Quadcopter Project


[FlorianH] has all kinds of new features to show off with this generation of his quadcopter project. Just about everything has seen an upgrade or some other kind of tweak since we looked in on the last version of the aircraft.

You’ll find some outdoor flight demo clips after the break. Right off the bat we’re impressed at the rock solid stability of the quadrotor while in flight. Even indoors the last version had a hint of a wobble as the control loop calculated stabilization. Here he borrowed some code from the open source Aeroquad project which helps account for this improvement. But the hardware choices lend a hand too. He moved from an ATmega32 up to an STM32F405RG processor. That’s an ARM chip which he programs using one of STM’s Discovery boards. The motors have all been upgraded as well (if you listen in the demo videos for both models you can hear a difference) and he redesigned the frame, which combines carbon tube with 3D printed parts to keep it light yet strong. The upgrade is every bit as impressive as the original build!

26 thoughts on “[FlorianH] Shows Off MinimaBL, The Next Generation Of His Quadcopter Project

        1. Try programming this in it will help:

          A robot may injure a human being or, through inaction, allow a human being to not come to harm.

          A robot must disobey the orders given to it by human beings, except where such orders would conflict with the First Law.

          A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

          Works for me :/

  1. Nice quad. Crappy music.

    I love quadcopters, want one, but I’m going to ask a very basic question: What are the advantages over a helicopter? I assume there must be some, but have never read anything on this topic.

    1. For me, quadcopters’ main advantage over helicopters is that they are mechanically really simple, just some sort of a cross frame with motors and props on them, no gears and stuff, so you have much more freedom on the design and less parts that can break at a crash (assuming your electronics are protected). Also, they tend to be more powerful (they have 3 motors more than helis, anyways) so you can lift stuff and don’t need to worry so much about weight as long as you don’t do aerobatics.
      They also are very stable, but I think good helis are, too.

    2. Quadrotors have really good maneuverability and can perform tricks that would be impossible for a helicopter to do. Depending on the quadrotor you can sometimes get a heavier payload on them.

      That being said, they’re really unstable and need computer-controlled stability to get them off the ground.

          1. Quadcopter wins! By about a million miles. The guy on the voiceover could get a job as a Speak-n-Spell, but that is awesome!

            One copter balances an inverted pendulum, maybe 60cm long. Then it throws it, and the other catches! Amazingly clever.

      1. No pitched main rotor = no auto rotate = no safe landing if a motor fails.

        Quads have 4 (or more) single points of failure, helicopters have one (or two if you count the tail rotor separately).

        Basically quads substitutes speed control on 4 or more fixed position rotors in place of a helicopters single pitch and speed controlled main rotor (which requires a tail rotor to maintain directional stability).

        Basically they are two very different birds (common factor is they both can hover and take off/land vertically).

        Can’t think of a “trick” that a quad can do that a regular helicopter can’t.

          1. I think the RC copter fans are getting defensive without actually thinking about what’s being said. A quad rotor can do things a more traditional helicopter can’t. It can also do things an F-22 can’t. That isn’t a value judgement; it’s just a fact. Different critters for different purposes.

        1. Really 4 single points of failure? I dunno about the software, but certainly mechanically it should be possible for a soft landing with 1 of the motors gone. Maybe even with 2 gone, if they were diagonally opposite.

          A quad with a failed motor would be a tri-copter, which certainly should be stable. Whether this is possible with the built-in equilibrium seeking of a particular copter depends on the machine. I’d guess that it’d come naturally, without needing the case coding for. But you could easily detect when a motor stops spinning, then automatically jump to a safe-landing routine.

          Most quads I’ve seen don’t even have gearboxes. Compared to the swash plate and other surrounding black magic, it takes a bit of thinking just to understand how an ordinary monocopter stays in the air.

          Come to think of it, auto-rotation wouldn’t be much to add to a quad. The fact nobody seems to have bothered might indicate that it’s not needed.

          1. quadcopter – 1 rotor = tricopter is a bit too easy… the center of gravity of the quad will stay in the middle of the quad, but the remaining 3 rotors will be off-Center leaving 1 quarter of the quad without thrust. even if the remaining motors would be powerful enough to hold the quad in the air, current quadcopter controls arent capable of controlling that and even if the option was there, i bet that would be far out of the PID regulation capabilities. Keep in mind that quadcopters arent aerodynamically stable, even to hover they need quite a bit computing.

            Autorotate with a quad is impossible. AFAIK helis autorotate by stopping the main rotor, meaning the heli is braking its fall by the rotor’s drag. As a quad can’t rotate around 1 central rotor (it has none), autorotate cant work.

            It’s logical that quads dont have gearboxes, there are no gears ;) thats one of the fascinations of quads for me, that there is, as you say, no “black magic” of choppers.

    3. Quads are easier to fly, cheaper to operate, require less maintenance(fewer mechanical parts) and take crashes much better.

      They tend to be very cartesian(flying in x,y,z) which feels a little strange, tri-copters are suppose to be more natural feeling, or at least more like a airplane.

    4. I have flown RC Helis for more than 12 years, and there is a world of difference between Helicopters and Quadcopters. First of all, flying a heli requires absolute concentration, it is like controlling a dinner plate on a queue stick, second, it is very difficult to learn how to fly one, and it takes a lot of practice to be able just to lift off. I was a member of a 130+ member RC flying club, and I was the only one flying one. Others tried, but had to give up. Flying a quadcopter is literally nothing to it in comparison. Also, a heli requires a very tedious pre-flight mechanical setup procedure, which can take more than a 30 minutes, and is very critical, for it determines if you will be able to fly it decently or crash right on take-off. None of that on a quadcopter.

    1. the STM32Fxxx have three bootloader spaces: CAN, UART, and user. It may also have another bootloader space for SPI (you can write a SPI bootloader for one of the boot spaces), but I am operating on memory without looking at the datasheet. There are two BOOT pins. I seem to remember if the pins == 00, it boots the application.

      Cool project. I like ARMS (I have two arms on an ARM project – my last I had 6 ARMS! – think about my name :^), but I am also an Atmel freak. I wonder what his footprint is – an ATXMega128A1 is a sweet powerful chip.

      Great project! Any link to the model he was trying to pick up? (wow. There is a pun in there somewhere…)

  2. First, I din’t think it’ll be able to fly with that super heavy landing foot.
    and secondly, although a magnifying glass is an interesting accessory, I’d be afraid of sticking my head into a propeller when trying to look through it. ;-)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.