Topics created by david.moro

@david-moro We have not used that specific radio before, only the nano and mini

@david-moro Yes, absolutely. if you are OK posting it here we can take a look or if you'd like to email the schematic to contact --at-- modalai.com we could take a look that way as well

@Moderator thanks for the information.

Could one of the two M0072 be used on the VIO pipeline at the same time? I am thinking it would be great to have VIO and DFS from just two cameras.

Hi @Trinadh

Can you confirm (using voxl-version) what version you have?

e.g. should paste this info:

-------------------------------------------------------------------------------- system-image: kernel: -------------------------------------------------------------------------------- hw platform: mach.var: -------------------------------------------------------------------------------- voxl-suite: --------------------------------------------------------------------------------

To confirm, you are using:
VOXL2 Mini -> J6 -> M0135 --> JL for ov9282, and JU for IMX214?

We don't have a binary setup for IMX214 or ov9782 on J6 upper, which is "ID 1".

ls /usr/share/modalai/chi-cdk/ov9782 com.qti.sensormodule.ov9782_0.bin com.qti.sensormodule.ov9782_2.bin com.qti.sensormodule.ov9782_3.bin com.qti.sensormodule.ov9782_4.bin voxl2:/usr/lib/camera$ ls /usr/share/modalai/chi-cdk/imx214 com.qti.sensormodule.imx214_0.bin com.qti.sensormodule.imx214_2.bin com.qti.sensormodule.imx214_3.bin com.qti.sensormodule.imx214_4.bin

So it seems like you would need a binary built for imx214 on "slot1", correct?

@david-moro , awesome! thanks for confirming.

Alex

@Alex-Kushleyev , I think that is plenty for now. We will order a couple and test them. Thanks a lot!

@david-moro Yes, the VOXL 2 Mini supports OV9782

@Aaky ,

Please check the following thread for more information: https://forum.modalai.com/topic/3459/masking-qvio/

My suggestion:

download the test mask from here : https://gitlab.com/voxl-public/voxl-sdk/services/voxl-qvio-server/-/blob/master/misc/mask_example_160x120.pgm . which has top half of image black (0), and bottom half white (255). values of 0 is where new features will be allowed to be detected You can use gimp, to modify the mask and before saving you can threshold it so that the values are all 0 and 255 (nothing in between, just in case). export from gimp as pgm (raw) do a test with the original mask example, make sure no features are picked up in the bottom image. This is the baseline test then try your new mask, you can also overlay the mask on top of the QVIO overlay, just like the other user did for checking.

Also please note (from the other thread) to pay attention to the fact that QVIO mask path is relative to where qvio server starts. So if you start it as a service, it's working directory will be / but if you run it manually, the working directory will be your current directory in the terminal - the other thread explains that issue.

Alex

@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?

@david-moro

Awesome! Glad we were able to get this resolved 🙂

Thinking about it more, this makes a lot of sense. A higher resolution camera is going to be ultra-precise in its pixel-to-world mapping meaning small inaccuracies in the grid would probably cause the model to blow up. I'll make a note to make this very clear on our documentation.

Don't hesitate to reach out if you have any other questions. Best of luck!!

Thomas