3D Robotics

3689404622?profile=originalWe're getting ready to launch the beta ArduCopter 2.0 (formerly known as ArduCopterMega, or ACM), and much has changed. First, it's our first real quadcopter UAV code, designed from the ground up for fully scripted autonomous missions as Jason's runs at Sparkfun demonstrated. It's also totally built on the APM code base, so sophisticated MAVLink mission planning and telemetry is built in and it inherits APM's mature navigation algorithms.


Replacing the LabVIEW configurator that we used in ArduCopter 1.x is a configuration utility built into Michael Oborne's fantastic APM Mission Planner. This both gives you command-line (CLI) access to the ArduCopter 2's interactive setup process, as well as a graphical display of all the sensors at work, so you can see at a glance that everything is working right. And because it's all MAVLink-based,  you can use it equally well on the bench via USB as at the field (or in the air) via Xbee. And it will work for any MAVLink-based UAV!


Unlike the old Configurator, you don't need to install LabVIEW to use this. Although the Mission Planner is Windows-only right now, you can use ArduCopter 2's built-in CLI to set up the board and software on any operating system, which will give you a text-based version of all the same data.  Our aim is to be totally cross platform, just like Arduino.


As with APM, the Mission Planner will also load the code, so there's no need for Arduino anymore (unless you want to tinker!). Our aim is to make ArduCopter the easiest quadcopter UAV setup on the market. We've still got a ways to go on that, but the progress so far has been very impressive. There are a lot of RC quads out there, which are increasingly easy to set up, but very few real UAV quads with autonomous mission planning. We'd like to make setting up a UAV no harder than setting up a RC aircraft. ArduCopter 2 is a big step in that direction.


As for the release of ArduCopter 2, we're very close. The dev team will review the last outstanding issues on Sunday, with the target of releasing the first beta next week, with documentation.




E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • I love the Mission Planner, it makes everything so simple for setups now. Another step I'd really welcome would be the ability to opt between auto update of the latest code from the internet, and also the option of a local folder to search for the latest code you may have downloaded. That way you can use Arduino to edit the code, and then point MP to upload it after. Doing all the hacks, compiling, and uploading is beyond my pea-brain I'm afraid. I'm trying to set voltage division ratio for the Current sensor, enable camera-stab and other goodies that aren't yet built-in to the optional hardware page on the setup in mission planner. I'm NOT complaining, it's just feedback from a grateful noob! ;)

  • oh we had some heavy cloud cover, also need to attach a battery to the gps for a warm start..

    just got back from using the copter in a competition, didnt do so well, but survived a water landing and managed to live!

    im updating to copter 2 code, xbee seems to be behaving oddly..
  • 3D Robotics
    Mees: Yes. ArduCopter 2 (formerly ACM) has all the same features as APM, including full two-way telemetry and mission planning.
  • will ACM also support xbee?
  • Hi guys

    Whats the E.T.A for arducopter 2? im not trying to sound pushy. I'm just really excited :)
  • Moderator


    What is telling you that you don't have lock? I was using the HK GCS and was beating my head for a lock until I decreased the GPS rate in the GCS...

  • summer will be fun :D .


    if i can get my gps to lock ...

  • Developer
    DCM works great on ArduCopter too. All codebases are based on that and after we tweaked it a lot it's even better now. Many other quad people has also adapted DCM algorithms.
  • Oh, i'm not saying it isn't working great, just that it might not be optimised for 'copters. If it is still based on mr. Premerlani's work, and if i've understood it correctly:

    1) it assumes forward flight
    2) it does not discern between gravitational acceleration and "translational" acceleration
    3) it doesn't really deal with translational drift (caused by wind, for example)

    None of that is critical, of course.
  • 3D Robotics
    The DCM is not "geared towards planes". It's a generic sensor fusion algorithm. It's linked to controls with PID loop, which are optimized for various platforms. We've been flying ArduCopter for nearly a year with it and it works great.
This reply was deleted.