Best posts made by Chad Sweet

Hi Cris, some answers to your questions:

1. Ubuntu 18.04 for the time being (datasheet).
2. ROS 1 and 2 are easily installed. Examples are pending. Current SDK is documented here
3. The image sensors are available via gstreamer. We will have an IMU API soon.
4. Soft AP and station mode are supported (docs)
5. VNC/GUI - we have not tried this, but it's Ubuntu so it seems likely something could be tried
6. See datasheet for architecture. Yes, there is an SD Card slot.
7. The kit comes with everything shown here, the 5G modem is optional.
8. The 4G add-on, 5G add-on, Microhard add-on are all optionally supported.

posted a few basics in the docs https://docs.modalai.com/voxl-tflite-server/

You can see exactly where the image comes in her: https://gitlab.com/voxl-public/modal-pipe-architecture/voxl-tflite-server/-/blob/master/server/threads.cpp#L105

Your understanding is correct, ModalAI's supported configurations are a result of Qualcomm limitations

The lens is fixed focus, but if you twist the little black part of the lens you should be able to focus.

It would have shipped with everything tested and working. What do you mean by "passed the wizard"?

Are you flipping the fastboot switch back? Your symptom is consistent with leaving the fastboot switch engaged instead of setting it back to normal after a few seconds

Might be worth watching the tutorial here if you're unsure

The system was tested in VIO mode for 20 minutes before shipping. All drones are very sensitive to frame modifications, so this needs to be done carefully

• Were you able to fly successfully before frame modifications?
• How well does it fly manually? Have you looked at the logs in manual.mode for vibration?
• Can you please share a log of the VIO output?
• If you reset all of the software, you may need to recalibrate the cameras
• It looks like there are loose parts hanging off of the drone. VIO is VERY sensitive to vibration. Can you revert your frame modifications and try again?

We do not have a thermal option available today, though the Flir Boson and Lepton have been integrated by customers in the past using USB

Stereo cameras need to be co-planar and mounted rigidly. Then they need to be calibrated. Otherwise you should be able to mount wherever convenient

For VIO use voxl-inspect-qvio

For GPS use voxl-inspect-gps

The source code for those is here

Thanks for posting the ssl fix. We're looking into this. We don't typically use rosdep for anything. Is there something specific you need to get working?

A more traditional ROS experience is easy to achieve running Ubuntu in a Docker on VOXL : https://docs.modalai.com/docker-on-voxl/

You can see an example of how to write telemetry on top of a video frame here
https://gitlab.com/voxl-public/modal-pipe-architecture/voxl-qvio-server/-/blob/master/server/main.cpp#L715

You should be able to use the open source PX4 avoidance, but it may require a bit of an atypical architecture. When trying to use big ROS projects we recommended running Ubuntu on VOXL in Docker

The base Yocto layer should be ROS core and expose the VIO and depth from stereo modules:

• voxl-cam-ros
• snap_vio
• depth-from-stereo

Then the Ubuntu OS in the Docker should subscribe to the appropriate hardware layer topics to execute the PX4 Autonomy

We should have this better documented, but keep asking on this thread and we'll try to get you going. Shouldn't be any major blockers, but a few tweaks are probably necessary.

We haven't tried the Avoidance stack in a while, so it's probably going to come down to what you are more comfortable with. If you are comfortable with Gazebo, SITL might be a good way to go.

The trick with ROS always comes down to making sure all of the nodes are talking to each other. There is a good tutorial on debugging ROS here

You'll want to have a list of the ROS topics the Ubuntu PX4 Avoidance stack requires, and then a list of the ROS topics the yocto base layer is providing. Then you'll need to map the two together.

You should only have one roscore, which should likely be in the Yocto layer. Your PC implementation should be able to see that if you have ROS_IP, etc configured properly.

The most valuable tool right now will probably be rostopic list and rostopic echo. Run ros topic on each of the layers:

• yocto base layer
• Ubuntu Docker
• PC workstation

Make sure you see the proper topics in each location and that they are publishing data.

Regarding stereo, I don't think that PX4 docs section is applicable. That RosDfsExample node should already be publishing a disparity map and point cloud. You can see it is publishing the disparity map here and point cloud here

It is the way for software to tell which TOF driver should be used. Setting to 1 supports the HAL3 method you are using, setting to 0 supports the older libCamera method using tof_cam_ros

The HAL3 method is intended to give the developer more control over the pipeline

Assuming the ipk is installed, can you please then run this command and then try the ROS node again?

setprop persist.camera.modalai.tof 1

We don't have a production mechanical solution yet for the TOF. We have used double sided sticky tape for prototypes, picture below. An extension cable, found here, might help make things easier.

You don't want to mount the fan with the board as the vibration will interfere with the IMU

Otherwise, airflow is what is important. We typically mount a few mm away, but precision isn't that important.

If the board is exposed while flying, the fan isn't really necessary. The fan is very useful for stationary or benchtop development

Definitely B, that is the Snapdragon you are trying to cool

When you go into fastboot, are you physically switching the fastboot switch back to normal? That switch needs to behave like a button (volume down on an Android phone). If the switch stays fastboot enabled, it will restart after 10 seconds

Even simpler to remember: https://code.modalai.com