"Exception thrown while processing service call: timeout” error
See original GitHub issueHello,
I am trying to use the crazyswarm repo to fly two crazyflies (for now) using VICON system. I have gotten to a place where our ROS system recognizes the existence of crazyflies. I verified this with the provided chooser.py script.
First issue: If I click reboot, I can observe the crazyflies switching off and turning back on, thus proving that the connection is established. However, if I click on battery, I see 0.00V for both of them though their batteries are charged. I’m not sure why reboot is working but not the battery option.
Second issue: I’ve selected only one crazyflie from the chooser.py. Before going to VICON, I want to fly the crazyflie with joystick. When I run roslaunch crazyflie_demo teleop_xbox360.launch uri:=radio://0/100/2M
, it throws an error saying “Exception thrown while processing service call: timeout”. No values were found in the IMU topic and we can’t see the crazyflie on the RViz display. The uri is correct as I cross-verified by running rosrun crazyflie_tools scan -v
. I think the connection is established as the reboot is working in chooser.py script. But I’m unsure on why I’m facing this error.
Any suggestions to solve these issues would be greatly appreciated. Thank you.
Issue Analytics
- State:
- Created 3 years ago
- Comments:16 (6 by maintainers)
@MauroPfister Thanks for the fix! Yes, adding that to the documentation might help many other people who are using the Crazyswarm on 16.04.
@manikamakam The errors you mention indicate that the tracking didn’t work. Usually the key question is if the tracking issues occur before or after the crash. One thing to note is that if you operate right at the limit of your mocap (e.g., markers too small, cameras far away etc.) tracking can be very difficult. The best solution in that case is to use single-marker-mode (with libobjecttracker) and put a single, larger marker on top. Make sure that whatever marker you use, that the marker appears at least as 3 pixels on each relevant motion capture camera. Also can you confirm that you verified the tracking and communication by logging
stateEstimate.x/y/z
?Sorry, not yet.
hit-JB @.***>于2021年12月6日 周一下午8:48写道: