Bitcraze Forums

Yes, that is possible. You can also you use the Crazyswarm with VRPN.

Estimator : In all recent firmwares the estimator is actually picked dynamically. If you have the flow deck installed, ill will automatically switch to the EKF. The compile time flag you use, simply forces the estimator to be always EKF (instead of dynamically chosen). Paper : That paper uses the C...

See response at the identical forum post.

I was able to reproduce this issue and it seems to be caused by an incorrect log TOC. As workaround, you can run rm ~/.ros/*.csv to remove the log TOC cache files and re-run again. In the future, it would be better to open an issue on github at https://github.com/whoenig/crazyflie_ros/issues, includ...

You need to set "enable_logging" and "enable_logging_pose" to True in your launch file (see https://github.com/whoenig/crazyflie_ros/blob/8a440ae4b43f4ce18dc4e1f7b4fceb7b2b3cec21/crazyflie_demo/launch/swarm_external_position_vrpn.launch#L19-L20 for an example that has "enabl...

Hi, which controller are you using (onboard PID, onboard Mellinger, or offboard crazyflie_ros controller)?

Point-to-point communication just means that the ROS package tries to ensure that packets can be exchanged between PC and each CF. If there is a timeout, an (uncaught) exception is thrown (and therefore the crazyflie_driver terminates). This is likely the desired behavior if you are flying 1-5 CFs, ...

Have you considered using the Crazyswarm instead? You are running into multiple issues here: 1) crazyflie_ros uses point-to-point communication and throws an exception if a connection is lost; 2) crazyflie_ros uses one thread per CF with the radio being a shared resource. The Crazyswarm mitigates th...

Crazyflie_ros is created and maintained by me (Wolfgang Hoenig). Crazyswarm is created and maintained by James A. Preiss and myself. We frequently contribute back to the official repos such as firmware changes, but we are independent researchers. Bitcraze maintains a Python SDK and client. Both Pyth...

I think just implementing BVC, ORCA, or Barrier Functions (see e.g., https://arxiv.org/abs/2002.11807) might not be enough for a MS thesis since that has been done. To my knowledge, all of those proofs-of-concept require a motion capture system though. There could be some merit to leverage the peer-...