This seems to be a popular ROS2 node that appears to be setup to use lidar: https://github.com/SteveMacenski/slam_toolbox Appears to do localization. We could probably attempt to modify the source code to send VISION_POSITION_ESTIMATE messages to the PX4 instance

@Caio-Licínio-Vasconcelos-Pantarotto No, voxl-mapper uses VIO for localization

@tom SOLVED, thanks Tom, that command line prompt showed me that I needed to voxl-configure-extrinsics. Thank you!

@david-moro That is my impression too, that and, it does not seem to be all that robust in general. I am very interested in using an open source VIO solution, and build it to be a mostly drop-in replacement for Qvio. I need it for high-altitude flight too, in mostly texture-less environments, a most robust solution seems necessary. I only just started looking into which methods could be interesting to use, and came across DM-VIO: Delayed Marginalization Visual-Inertial Odometry which seems to be very modern and robust. The paper for this states that it was run in real time on a 2013 mobile i7 at 2.3GHz, so I am hopeful that it could run on the VOXL 2. I would be interested in following along if any of you decide to go in another direction than Qvio.

@Moderator will try to on our next flight. Do you have an easy way to trigger a capture when certain conditions are met? It is hard to correlate hundred of images to the autopilot logs to find these instances. We are working on streaming the qvio-overlay as an RTSP stream down to the GCS. That might be easier. Unrelated question. Can the VIO algorithm use the colored images coming from the camera? If yes, is there an advantage at all?

Thank you. I got the VIO going on the drone. I noticed that when running the VIO server, the IMU_BW error kept coming up and that was the cause of the server's crash. Therefore, I increased the limited_imu_bw_trigger parameter in the voxl-qvio-server.conf arbitrarily to 100 and it worked fine.

You can't really tune the feature detection algorithm. You can tune the exposure algorithm. You might try to increase ae_msv parameter in camera-server config to maybe add more contrast?

@Steve-Turner Thank you for your help. I realize you are not part of the ModalAi Dev Team, but I would like to ask for your help with this. There is more information that I have posted on this thread. I am lost, and would really appreciate any kind of help you could provide. Thank you.

@Steve-Turner - Yeah I have put EKF2_AID_MASK to 0. Then the drone would just be behaving normally. @Cliff-Wong - I understand that I would have to disable qvio-server, if I would only fly in non-evdata modus. But when flying with evdata, if then the evdata goes bad, I need to switch to non-evdata to be able to save the drone from crashing. But now the drone will still crash!

Hi @Matt-Turi, Thank you very much for your clear answer. We will give this a try! If you have any examples or more information on how to do this, please feel free.

Thank you for your response Matt

Anything with visual texture will improve performance. It does not have to be an AprilTag. Just anything with a non-repeating pattern

Hi, The MV library is made by qualcomm and has some licenses attached to it, we can't publish (and don't even have access to ourselves) the source code and have to keep the headers behind a EULA. Since both our voxl system images and docker build environments are behind said EULA, that's where we keep the header. It can be found in /usr/include/mv/ on both the voxl and the voxl-emulator docker image.

@darwin-keem The map is published in real-time by voxl-portal to a websocket at: ws://<VOXL_IP>:80/mesh. If you would like to process the map real time without reading from disk, you could have another service connect to the socket and parse the binary data as defined here in voxl-portal: https://gitlab.com/voxl-public/voxl-sdk/services/voxl-portal/-/blob/master/web_root/js/mapper.js#L260

@Matthew-Booker thank you!

@Hyosun-Yoo said in [yocto:/# roslaunch voxl_cam_ros tracking_slave.launch] Error Message ?: Re: AprilTag Relocalization Dear I got an error message when I run this command. [yocto:/# roslaunch voxl_cam_ros tracking_slave.launch] Please tell me how to solve this message. VOXL Vision (modalai-3-2-0-0.3.4-b.tar.gz) VOXL Vision PX4 Apriltag Relocalization error Error message ERROR: could not open camera subscriber for cam id 1 [ERROR] [1034.046473200]: CameraManager::Initialize() failed [ERROR] [1034.046845023]: Unable to open camera. ===============* [image: 1660694916409-5c2c6beb-9c8b-432b-8430-fcc4d81c1ef3-image.png] Thank you so much. Mr. Yoo, hyosun

Hi, There are two answers to your question: You can pipe depth input from voxl-dfs-server (if you're running stereo cameras) to voxl-mapper, or any other service that you may create that produces depth estimates (i.e. if you had a depth from motion project). We absolutely plan to support the tof camera on voxl2, we're working on this right now internally and hope to support it in the next month or two.