A number of small issues

Post here to get support
Post Reply
ArmchairArmada
Beginner
Posts: 2
Joined: Fri Nov 18, 2016 3:37 am

A number of small issues

Post by ArmchairArmada »

Hi, we bought a few Crazyflie 2.0s with the intention of using them for education, but I am having difficulties getting started with using them. I am using the virtual machine in VirtualBox.

* The first problem is that often multiple Crazyflies are detected even when only one is powered on, so the example code often selects a "phantom drone" instead of the real one.

* The second problem is that sometimes the client gui connects to the crazyflie and shows battery power, but none of the sensor input updates (the flight controls remain unchanged).

* Third is frequent disconnects -- more often when the drone is a little over ten feet away.

* Fourth, when trying to use an XBox 360 controller, the drone frequently becomes momentarily unresponsive (lost packets, I believe.)

* Fifth, in general, I have been having difficulty finding comprehensive documentation about the drone and how to program it -- I am aware of the Wiki, but it does not seem to have information on everything (as far as I can tell).

The Crazyflie 2.0 seems like a good product and I have seen some impressive things done with it, but I am having a hard time overcoming these difficulties. Thanks.
arnaud
Bitcraze
Posts: 2538
Joined: Tue Feb 06, 2007 12:36 pm

Re: A number of small issues

Post by arnaud »

Hi, Sorry that you are having trouble getting started.

The Crazyflie comes programmed on channel 80 datarate 250k (radio://0/80/250k). If the crazyflie is close to the crazyradio we have seen cross-channel interference. You should always connect to the one on channel 80. To fix this problem you can set your Crazyflie to 2M datarate, this way the ghost channel will disapear and the Crazyflie will also be most resilient to interference coming from Wifi and Bluetooth. Depending of your environment changing the channel can also help.

The four first point seems to be related to bad radio communication, maybe by connecting to one of the "ghost" channel instead of the right one. So fixing the radio communication should solve those points.

For the fifth point, there is a getting started with development documentation that goes throw making a modification to the crazyflie and flashing it in: https://www.bitcraze.io/getting-started ... velopment/. This is a good starting point to understand the process. Could you maybe tell us a bit more about which documentation you are missing. Being a development platform the code is always evolving so it would be hard to very precisely document everything, but there is for sure a pretty good margin for improvement.

/Arnaud
ArmchairArmada
Beginner
Posts: 2
Joined: Fri Nov 18, 2016 3:37 am

Re: A number of small issues

Post by ArmchairArmada »

I tested out a few things -- it seems I have to give up on trying to use an XBox 360 controller. It did not seem to be radio interference, since it did not matter what channel was being used. After simply not connecting the XBox 360 controller to the computer things seemed to start working more consistently. I also increased the data rate to 1M, which might have helped.

As for the information part, we were intending to have students program them in Python but I could not find much information on how to get the Crazyflie to fly in less erratically -- like by using sensors maybe? We have not purchased the positioning system due to budget and wanting to minimize setup for the student using them at home (they'll only be using it for around two weeks). The problem is, every pre-programmed test flight I try running (just lift off, attempt to hover for two seconds, then land) result in the drone crashing into something. We might not need precise controls (as it is an introductory thing to get students interested in programming) but it would be nice if it could stay within at least a three foot radius for a five second flight. I tried adjusting the pitch/yaw settings but it didn't seem to significantly help.

What would be really nice for the students would be a tutorial series with activities they can follow along with. Other schools seemed to have been using Crazyflies in education, but I could not find much information about the lesson plans they were using.
whoenig
Expert
Posts: 395
Joined: Mon Oct 27, 2014 2:55 am

Re: A number of small issues

Post by whoenig »

All the issues you described initially seem to be related to bad communication. You can use https://github.com/bitcraze/crazyflie-l ... io-test.py to find a good channel. I also strongly recommend the 2M data rate.

I don't think the autonomous flight as you want it is possible with the Crazyflie without additional sensors (such as optical flow sensors) or external sensors (UWB or camera). I sent you a PM with some more pointers regarding education.
theseankelly
Expert
Posts: 153
Joined: Mon Dec 28, 2015 3:23 pm
Contact:

Re: A number of small issues

Post by theseankelly »

I had a problem a long time ago that sounds very very much like what you're describing -- especially the range/disconnects/phantom device detection.

I ended up having the wrong firmware flashed on the CrazyradioPA: I had built for the Crazyradio and not the Crazyradio PA. Try updating that firmware. When you build it, make sure you use the command "make CRPA=1" as described in the readme: https://github.com/bitcraze/crazyradio- ... /README.md

Edit: you might also try it out from the mobile app to try and isolate the radio/PC as the problem.
http://www.thejumperwire.com
Tips, tutorials, and science about DIY electronics, drones, and embedded software.
Post Reply