@WesleyHavocAI Thanks for catching it! We just updated flat builds and now it contains m0054-2!

Regarding the feature track tab, I recommend upgrading to the latest available SDK, features should show on voxl-portal. The correct overlay is under VIO indeed

@tom got it, I think you indirectly answered my question by giving me the last valid configuration. Am I looking in the right place?: voxl sdk 1.5.0 -> system image 1.8.04 -> recipes-kernel/linux-msm/files/dts/common/m0xxx-modalai-gpio.dtsi : &soc { voxl_gpio { ... modalai,gpio-init-output-high = <0 1 53 56 57 85 86 87 88 89 124 152 157 159>; modalai,**gpio-init-output-low** = <36 37 **46** 54 55 67 82 83 84 131 153 154 155>; modalai,gpio-init-input = <48 49 50 51 52 110 111 112 113 114>; status = "ok"; }; ... }; So settings for gpio pin 46 on sentinel w/ spektrum receiver should be direction:out, value:0 assuming they aren't overwritten somewhere else.

@PatkirkMartin No, as I said a UART mapped to the applications processor would also work. PX4 has a split architecture on VOXL2 with all flight critical operations happening on the DSP and support functions such as Mavlink, logging, and data manager are on the applications processor. But there is flexibility to move functions between them as needed. So, to match the ELRS receiver with the Mavlink module you could run it on the applications processor instead of the DSP. However, you would need an expansion board that exposes such a connector.

@LukeNow This worked, thanks.

@mkriesel It's possible the kernel you're flashing is incorrect. Can you provide the full terminal output of the install?

Hi Cbay, There is a small clock buffer on the edge of the PCB. An RMA can only replace the whole board, not replace a single device. The part to check would be here: [image: 1779297839725-image-1.png]

Thanks @Eric-Katzfey ! Here is the pinout for the dev board "UART" port that includes i2C_SCL and I2C_SDA. [image: 1778773646212-0cda71eb-2c0f-4b1b-ba91-8b3862fc9763-image.png] The dev board pins 1-6 matching to J19 pins 1-6 We will give the COMPASS_DISBLMSK parameter a shot and see if that works.

@Plingaas02 Regarding odometry in voxl-mavlink-server, it will not modify timestamps on this message. There will be, potentially, VIO odometry messages going to the flight controller from voxl-vision-hub via voxl-mavlink-server and flight controller odometry messages being routed out to the GCS and to the onboard stream. The odometry messages coming from PX4 will have the DSP monotonic timestamp which is different than the Linux monotonic clock and these will drift with respect to each other over time. PX4 runs the timesync service so it can correct the incoming messages from VIO.

Hi @austin-c Sorry I missed this message somehow. I will email you to kick this off with your team.

@austin-c , The GPS driver is built for SLPI as well as APPS PROC, so you could connect it to a apps-proc UART and start the driver on the apps proc side of PX4. This was done on RB5 : https://docs.modalai.com/Qualcomm-Flight-RB5-linux-user-guide/ https://docs.modalai.com/voxl2-linux-user-guide/ Regarding magnetometer, just like with the GPS driver, you should be able to enable the mag driver to build for apps proc. Since the apps proc i2c uses standard posix interface, it should just work. However, i don't think we have ever tried it. Alex

@restore There are a few examples of code that listens for Mavlink messages in our SDK. For example, in voxl-mpa-tools, take a look at the source code for voxl-inspect-mavlink.c. https://gitlab.com/voxl-public/voxl-sdk/utilities/voxl-mpa-tools/-/blob/master/tools/voxl-inspect-mavlink.c?ref_type=heads Probably would be a good idea to run voxl-inspect-mavlink on the mavlink_onboard and mavlink_to_gcs pipes to see which one has the desired Mavlink message in it. Then create your own application that listens for the Mavlink message on that pipe.

@cbay voxl-fpv-px4 is based on voxl-px4 but has some custom tuning code / parameters for our fpv drones that isn't in voxl-px4. So, it's kind of a superset. But there isn't anything related to low level i2c drivers so I have no idea how that could be causing this issue.

Hey @svempati I trained a model in the docker command as provided by you. I am able to run an inference on potholes following exactly the steps provided by @Zachary-Lowell-0 earlier. Could I ask you what your voxl-tflite-version is? And if you havent yet, could you try to run this version: http://voxl-packages.modalai.com/dists/qrb5165/sdk-1.6/binary-arm64/voxl-tflite-server_0.5.1_arm64.deb Make sure to change the .tflite model and labels as shown in the loom video after the install, then run voxl-configure-tflite Thanks

@Moderator Thanks for the clarification on the ~50 ms glass-to-glass number. Trying to get a better understanding the performance envelope: 1. Does the 50 ms include RF transport over Doodle, decode, and HDMI output latency, or is that measured only within the VOXL + VRX software stack? 2. Was this measured in point-to-point mode or mesh routing mode? 3. Can you expand on bitrate, resolution, and encoder settings were used (GOP structure, B-frames disabled, UDP vs RTSP)? 4. Is MAVLink or other traffic running on the same link during that measurement? 5. Do you have a latency breakdown per stage (encode / RF / decode / display)? Appreciate any information you can provide. I just want to make sure I can test and validate this effectively.

Hi @wetherbeej Oh yikes.. You should really work with us on that change. There is a high likelihood to damage your VOXL 2 and the MEMs sensors are likely all broken now from the extra reflow cycles. Flash provisioning is all done over QFIL tools, not PCAT.