Posts made by david.moro

Hi, we are trying to spin a flex cable for a tight integration in a module and could use some help. Would the ModalAi team be able to have a quick look at our schematic?

Thanks!

@Alex-Kushleyev, no need for super low-latency FPV, just using a higher megapixel sensor to stabilize down to what we can consistently stream down the radio link. Some form of digital gimbal would be great too, where we can steer the field of view to a specific region

@Alex-Kushleyev , we would love to use this feature too. Please keep us posted, and feel free to reach out if you need beta testers!

@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.

I came across this SKU earlier today: https://www.modalai.com/products/m0072?variant=40414007132211

I have two questions. First, can this sensor substitute the standard tracking camera? It looks like the specs are the same. And the shorter flex cable would be very helpful in one of our integrations. Secondly, can you share any details on how to use it for stereo depth? How does it perform compared with the standard camera? How hard is it to calibrate?

Thanks!

@Nathan-Raras . Understood. Thanks. At this point we are ideating/prototyping, so please don't waste much time on it. Knowing that it is possible will help us lock in the camera configuration we will use in the final design. Cheers!

Understood. Thanks. Is there a path to add support to the ov9782 and the imx214 simultaneously on J6? Just curious. We are trying to integrate on a very tight space, and connecting to a single camera port simplifies the mechanical assembly.

Hi,

I am trying to get a config running with the ov9782 (M0113 + m0008) and a imx214 (M0025-2) camera connected to a dual sensor flex (M0084) on J6 of a VOXL2 mini. It seems like the ov9782 folder /usr/share/modala/chi-cdk does not have a driver for that camera on sensor HW ID 1. Is it possible to get this driver compiled and added to the sdk? Thanks!

David

@Alex-Kushleyev , it worked. Thanks!

Awesome! Will give it a try. Thanks @Alex-Kushleyev !

Hi, I am working on the integration of a camera on the VOXL2. The camera acts as a USB CDC device and outputs an 8-bit greyscale image on ttyACM0. The manufacturer provides a python example to talk to the camera. I was able to deploy it on the voxl2 and read the images. Now I would like to push those to a pipeline for consumption by voxl-streamer and other processes. Do you have any sample code or guidance on how to do this on python? I would like to have a functional prototype of the whole thing before porting it to C++.

Thanks!

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

@Moderator , @Alex-Kushleyev , thanks a lot! Please keep us posted!

@Moderator, do we have any more specs of this module? Weight, power? Also, is there an ETA for it to be in production?

Regards,

Hi @Moderator, we are looking to integrate the ToF sensor in one of our drones and we just found out it is EOL. According to the product page, you are working on a replacement. Would you be able to share any details on it? Is it larger/smaller? heavier/lighter? Does it have similar range? Thanks!

@SMRazaRizvi We did not. But the more I look at the qualcomm API for mvVISLAM the more I think there is aren't very many knobs to turn.

We successfully installed a VOXL2 and OV9782 for tracking on a vehicle. When trying to replicate the setup on a second vehicle we found out that the image is upside down. The mechanical design does not have room to rotate the sensor. Can this be done in software?

Thanks

@Moderator. en_auto_reset is true for these flights. What we see is that once a reset kicks in, it has a hard time staying stable. My question above is more along the lines of how to make the vio algorithm not drop points so aggressively. That is what the images above show.

Cheers!

@david-moro . Quick follow-up.

We are doing testing up to 400ft with moderate success. One thing we notice is that the qvio algorithm can struggle to converge after a reset. I attach two picture. On the first one, right after the reset, there are 25 points being tracked.

A few seconds later, this drops to 3 points. The vehicle barely moved between frames.

A few seconds after this the qvio reset again. This behavior is very common once the first reset happens. Up until that point, if we take it easy, we can fly at 400ft without any issues. Could this be related to mistuned parameters? In particular, what is the effect of cam_noise_std_dev, min_std_pixel_noise and fail_high_pixel_noise_points? Can those be used to trade robustness for accuracy?

Thanks!