6 Replies Latest reply on Aug 7, 2018 7:30 AM by Intel Corporation

    Can't get coav-control working

    jonathanc

      Hello,

       

      I am working on an Intel RTF Aero drone. I am testing the Realsense 200 with collision avoidance to determine if it would be a good fit for our products. I have run into a lot of trouble.

      I am running Ubuntu on the Intel Aero.

      I am able to build the coav-control executable using the instructions on GitHub - intel/collision-avoidance-library: A framework for testing and benchmarking collision avoidance strategies

      The problem is when I run coav-control it does nothing.

       

      I ran the following command to start the executable

      ./coav-control -d DI_OBSTACLE -a QC_STOP -s ST_REALSENSE

       

      I then get the following output:

      Using Detect Algorithm: DI_OBSTACLE

      Using Avoidance Algorithm: QC_STOP

      User Sensor: ST_REALSENSE

      [mav_vehicle] Our system id: 2

      [mav_vehicle] Waiting for vehicle...

      [MavQuadCopter] mavlink_vehicle instantiated

       

      No more output is displayed.

      For me to get this much working I had to start the coav-control process on boot by modifying the rc.local script. I then piped the output of that process to a file so I could see what it was doing.

      The drone never attempts to avoid obstacles.

       

      What am I doing wrong here? Any suggestions on how I can get this working?

       

      I have also been a little confused by the instructions in the repository starting at "Deploying on Intel Aero"

      Those instructions are designed only for those who are using Yocto however all the forums and documentation I have read strongly recommend installing Ubuntu on the drone. In fact most of the setup instructions and commands are Ubuntu-specific. When I first started this project I tried to get Yocto to work with this library but I ran into a lot of headaches because Yocto has a light version of Git and every install command I need must to be modified to work on Yocto.

       

      Any help would be great! We have a lot of custom drones we are developing and it would be awesome if I could show the CEO a drone with a working obstacle avoidance system using Realsense.

       

      Thanks,

       

      Jonathan L Clark

      Software Engineer

      xCraft