@Alex-Kushleyev Here are the results of testing with a brand new GPS from a different manufacturer (https://holybro.com/products/micro-m10-gps?variant=42981482954941). The GPS is in the same position as the original, but connected to u-center. As with prior tests, Lepton is disconnected and ALFA is removed.

Top u-center screen shot is with drone powered OFF, bottom is drone powered ON. To me this Holybro GPS looks about the same as the 3DR GPS, i.e. satellite signals drop significantly with drone powered ON. Not shown are results where I found that the satellite signal increased significantly when I moved the GPS module farther away and/or added shielding.

What do you suggest as next steps?

Thanks for your help.

Holybro.jpg

holybro-gps-drone-off.png

holybro-gps-drone-on.png

@GlennTee Good news! The program now runs, turns out I needed to add the .py extension to the file name when calling the run command, so it should be "ros2 run px4_ros_com offboard_control.py"

However, we experienced a lot of errors while running. The drone did arm (But only would do every even numbered attempt for some reason), however it did not fly. I'm working now to debug these errors.
image(1).png

@brianaustin great to hear, thanks for the update!

@Eric-Katzfey

The SLPI image used to be part of the main system image. It was then separated out into it's own package for easier maintenance.

Gotcha makes sense!

So you have a very old version missing many important bug fixes. We've never tried installing the latest modalai-slpi package on an old SDK but I think it will work. Give it a try and see what happens.

Will do, just wanted to confirm that it "might work" so not totally wasting my time exploring this avenue haha.

But, obviously, it's really hard for us to support you when you use such old software with custom modifications on top of it. You should really try to make any customization such that they can easily be used with newer versions of VOXL SDK as they come out.

Yes this is something that we constantly run into at Cleo as a small company trying release a stable product but also keep up with the latest and greatest from modal and other open source vendors. As a dev at Cleo it feels like two things are pulling in opposite directions:

Having your base platform constantly updating, which then requires patches for API changes and sometimes more low level incompatibilities that come along with those updates. Getting the stability and functional improvements that come along with those base platform updates.

I'm sure these competing forces are felt by others as well, not just at Cleo. It's a conversation that is perhaps worthy of a call between Cleo and Modal devs to get aligned on the best way to get the stability and functional improvements from each vendor (modalai) release while at the same time minimizing the Cleo dev time needed to do those API patches and minimize incompatibilities or at least forecast them before spending the time trying to do an upgrade and then finding a incompatibility.

@voxl-phoenix , I think you also mentioned that you tried swapping the cables between tracking and imx412 camera, right? If so, it seems something got damaged in the IMX412 camera during connecting / disconnecting. For disconnecting the cables, we suggest using a special tool, please see https://docs.modalai.com/micro-coax-user-guide/#-proper-cable-and-adapter-handling- .

Also, it is hard to say for sure where exactly the issue is - M0173 board or IMX412 camera. the only good way find the cause is to replace components one at a time.

Just to elaborate on the symptoms that you are seeing : voxl-camera-server -l command essentially probes all the cameras (based on the sensormodule files located in /usr/lib/camera). Probing involves turning the camera on via the reset GPIO and attempting to talk to the camera via I2c interface. If the camera responds, the probe is successful, otherwise failed. In your case, the probe of IMX412 failed, which means either the camera could not be turned on for various reasons or the i2c communication was not working.

If you have another tracking camera AR0144, you could connect it instead of the IMX412 and see if the probe succeeds. You would need to add /usr/share/modalai/chi-cdk/ar0144/com.qti.sensormodule.ar0144_1.bin file to /usr/lib/camera (which is the AR0144 sensormodule for the camera slot 1, where the IMX412 is currently connected). Then if you run voxl-camera-server -l and detect two AR0144 cameras (3 cameras total) then i2c communication and power should be working).

Alex

@kestrelsystemsbp Perhaps you can do a fresh installation on 1.3.5 and see if the problem goes away? v1.3.5 has a lot of test time on it and nothing like the issue you are seeing has been encountered by us or any of our customers so this is a bit of a mystery. I would try the fresh install and, without making any modifications at all, see if it comes up and stays up. If so you can start introducing any configuration changes you need one by one to see if any of them are causing the issue.

@claw Not at this time as ROS2 takes up a lot of space so we don't include it by default

@Alex-Kushleyev understood,

Thanks again I will give this a try - fingers crossed

@SKA voxl-vision-hub will print out debug messages showing VIO status as it is running. Can you see anything that may indicate why it thinks VIO is failed?

@Raza-Rizvi In that diagram it shows PWM out to the ESCs. Are you planning to use PWM ESCs or the UART based ModalAI 4-in-1 ESC?

@Alex-Kushleyev
I understand. Thanks for the quick reply.

Thanks,

JS, Kim

@Alex-Kushleyev Can you provide me an EMAIL where I can share the video?

@griffin , please see the following post with the same symptom and fix : https://forum.modalai.com/topic/4171/frame-size-mismatch-got-43200-bytes-from-pipe-expected-38528

We will test this and merge to dev soon, meanwhile if you can re-build the camera server, you could test it.

Alex

@rdjarvis Okay, great! Instead of starting voxl-px4 with systemctl start voxl-px4 can you start it as a foreground process using voxl-px4 -d and then capture the output and paste as a response? I'd like to see if there are any error messages printed that could help determine why it isn't seeing your GPS unit.

@Alex-Kushleyev We're using the "hires_front" camera on the Starling 2 Max. I don't know exactly what type of camera (hardware-wise) is on the drone, nor how to find that information, but this is the current configuration for that camera in the voxl-camera-server.conf file:

{ "type": "imx412", "name": "hires_front", "enabled": true, "camera_id": 1, "fps": 30, "en_preview": false, "preview_width": 640, "preview_height": 480, "en_raw_preview": false, "en_small_video": true, "small_video_width": 1024, "small_video_height": 768, "small_venc_mode": "h265", "small_venc_br_ctrl": "cqp", "small_venc_Qfixed": 30, "small_venc_Qmin": 15, "small_venc_Qmax": 40, "small_venc_nPframes": 9, "small_venc_mbps": 2, "en_large_video": false, "large_video_width": 4056, "large_video_height": 3040, "large_venc_mode": "h265", "large_venc_br_ctrl": "cqp", "large_venc_Qfixed": 38, "large_venc_Qmin": 15, "large_venc_Qmax": 50, "large_venc_nPframes": 29, "large_venc_mbps": 30, "en_snapshot": true, "en_snapshot_width": 4056, "en_snapshot_height": 3040, "exif_focal_length": 3.0999999046325684, "exif_focal_length_in_35mm_format": 17, "exif_fnumber": 1.2400000095367432, "ae_mode": "isp" }

Does that answer your question?

@Maanav-Iyengar , you can try to move the camera to J6L and use sensormodule with ID 0.

If that does not work, then the issue is most likely not in VOXL2, but somewhere upstream (M0135 or farther up)

If you have another camera, you can try plugging it in instead of IMX412, so this would test the M0135 + M0036 path. Any camera would do, but you would just need to set up the correct sensormodule .

If you use any other camera for testing, which already has a flex on it (unlike M0107 IMX412), then you can skip the M0036 extension flex (you can test with and without it). If you use tracking camera, keep an eye on pin1 orientation, as it is backwards compared to hires cameras.

Alex

@Suvasis-M https://docs.modalai.com/voxl2-mini-external-flight-controller/

Hi @david-moro ,

I made several videos that show the current functionality of the ROI and EIS features.

ROI features: https://www.youtube.com/watch?v=FXv4855WjNc EIS: https://www.youtube.com/watch?v=fi2BO_U5f-c EIS outdoor test : https://www.youtube.com/watch?v=oHDsiTD8Ky4

Please note that the last video was recorded with different EIS settings than currently are used as defaults. However, the same processing pipeline was used.

We will support up to 4K60 stabilized output and concurrent EIS on multiple cameras (already tested). There will be a lot more work done in the coming weeks to improve the video quality when testing on actual flying vehicles.

Alex

@Samuel-Lehman I think the issue is with the tflite version running on voxl, which does not support v3 ops. An update to tflite on the voxl is being pushed but it will take some time to be merged into the main SDK. However, you can try tensorflow 2.17.1 from this branch.

Note that since it is not completely integrable, running it is slightly more complex. Make sure you are using the newest voxl-cross image to build voxl-tflite-server.

On Voxl, it cannot run directly, and you would have to run it via a Docker container. You can find a sample Dockerfile provided in docker_samples/ in the 'update_tf' branch. For further context on integrating opencl please refer to this.

Hopefully, the updated tflite will contain the necessary v3 op and it works smoothly.

@Alex-Kushleyev

I tried your advice, and I'm curious about the messages which looks like hardware related:

[ 6.363892] qcom,camera ac4f000.qcom,cci:qcom,cam-sensor1: Dropping the link to regulator.79 [ 6.413925] qcom,camera ac50000.qcom,cci:qcom,cam-sensor2: Linked as a consumer to regulator.79 [ 7.591724] CAM_ERR: CAM-ISP: cam_ife_csid_cid_reserve: 1040 CSID:2 IPP resource not available

Should I check the physical connection between the camera and the board?

I also give you all I got below:

Screenshot from 2025-03-28 16-57-06.png

Screenshot from 2025-03-28 16-58-12.png

dmesg log
https://drive.google.com/file/d/1SR9mGPbrsWaXHDW1HW4sHUVJHjCLTbcf/view?usp=drive_link