ModalAI Forum

Hi ZBFT, We have done the steps you recomended both with 1.6.4 alpha2 version in with 1.6.6 version. canceling qvio, making sure only one open-vins service regarding vio. recalibrating imu. The drone is still loosing horizontal pos at the same manuver( sliding to the left when there is a wall in front) I want to proved you with some facts. we reverted back to a code we know worked well before and suddenly we experiance this problem. ( made us syspect somthing changed in hw ( some kind of movment of camera or lense because of a crash) the qvio with 1.6.4 alpha2 version with qvio was working perfectly with this version of ours and now it not failing in the same manuver again and again. we double check all px4 params against a golden copy. The other possiblity that it is something wrong in our appication is also a possiblity( we have two drone acting the same) can you guide us what other information we can log in order to trace why vio pos suddenly jumps to a far horizontal position. what can explain such a behavior ? BR, Roy

Hey @Eric-Katzfey , I have similar query. Not exactly the same, but I want to run ardupilot along with Voxl fpv 4 in 1 ESC. Ardupilot on 3rd party FC. The architecture that I would like to try out is : Voxl 2 ---> 3 rd Party FC (with ardupilot)----> voxl FPV 4 in 1 ESC Especially since flight core v2 has reached EOL. Is any new FC being release? Please help me out here.

@ModalAI-Team Help me out here . I want to purchase some new components, but with out having a clear idea on the architecture I won't be able to do it. Please guide me here.

@ModalAI-Team If anyone could help me understand this behaviour and help me solve the boot issue, if there was any way, would mean a lot. I have 2 VOXL 2 which are not booting. which gave successful result in unbricking but did not detect ADB earlier. That was until last day when while trying unbricking the same boards, I got a positive result in one of the boards. If this issue can solved by anyway, please help me out.

Hey @Moderator , I had few more questions regarding the same. Does the 3D map generated have the same quality has PMD TOF? Can the 3D map be used for P-to-P navigation from voxl-Portal? Can the TOF data be used for collision prevention? Can multiple maps be used together? Does Multiple TOF give collision prevention in multi direction? If you could provide a image of map generated by the new TOF, it would be great and help us to make a decision for our architecture. Thank you for your help and time, looking forward to your response.

Hello, I would like to follow up on this topic. We are currently evaluating the feasibility of using Starling 2 for indoor autonomous flight applications in Japan. Understanding the recommended Wi-Fi configuration for operation in Japan will help us continue our evaluation. If anyone from the ModalAI team could provide guidance, we would greatly appreciate it. Thank you very much.

I am able to download the log from the site along with all of the parameter settings. The accelerometer bias is clear in the log. Can you try to recalibrate the IMU with air flow on the unit to keep it from getting too hot? The yaw estimate error is not active during the log. However, to remove the interaction with VIO, try setting EKF2_EV_CTRL to 0 and see if that clears that occasional error. If it does then that helps narrow the issue to the VIO input versus the magnetometer / IMU input. Is the red light on the GPS unit or the RC receiver?

@Alex-Kushleyev Hi Alex, poking into stereo a little bit. Why did you mention that guaranteeing extrinsics not changing might be hard? If hard mounted why would it ever change? Vibrations? Just curious. Because also couldn't this same logic also apply to just a one camera system? Thanks!

DSHOT Telemetry seems like the only option for the MadMotor ESC linked in the original post so that would be our requirement.

MCBL-00084-180 (26-pin Micro-Coax Cable): Operating temperature 0°C to +70°C; Storage temperature 0°C to +70°C. MSU-M0161 (EO Camera): Operating temperature 0°C to +70°C; Storage temperature 0°C to +70°C. (The Sony IMX412 image sensor guarantees image performance from −10°C to +60°C.) VOXL 2 / MDK-M0154-1-00 (Flight Computer): Operating temperature 0°C to +70°C; Storage temperature 0°C to +70°C. MDK-M0155-1-00 (Image Sensor Micro-Coax Adapter): Operating temperature 0°C to +70°C; Storage temperature 0°C to +70°C. MDK-M0062-3-00 (VOXL2 Ethernet & USB Hub Adapter): Operating temperature 0°C to +70°C; Storage temperature 0°C to +70°C.

Hi Alex, Thank you very much for your detailed explanation. It was very helpful, and I really appreciate you taking the time to investigate this. We understand that the OV7251 camera has reached EOL, and we are also aware of the AR0144 solution. We are planning to evaluate the AR0144 cameras in the future. However, for our current project, we need to reproduce our existing system using the hardware that we already have. Therefore, our immediate goal is to implement the following configuration: * 4 × OV7251 cameras * 1 × Boson+ connected through the M0181 VOXL2 MIPI Boson Adapter on J8 This configuration is very important for our current project, so we would like to continue with the OV7251 cameras for now. Once you verify the kernel/device tree changes, do you expect 4 independent OV7251 cameras plus one Boson (via M0181 on J8) to run simultaneously under voxl-camera-server? Thank you again for your support. We really appreciate your help, and we look forward to hearing your results. Best regards,

Hello @jameskuesel , Sorry for the confusion about the lens change. Please see the details below. The original lens in the IMX412 modules (M0107 and M0161) had manufacturing issues where the internal components of the lenses were loose and vibrating in flight, causing shaky video. Due to quality / vibration issues of the original lens (part number M10000580) shipping with IMX412 (M0107 and M0161) modules, the default lens was changed to M10001028 (late 2025). However the new lens does have much narrower FOV. When purchasing the standalone camera, you have a choice of lens: M0161 with old lens (wide) : MDK-M0161-1-07 (3.1mm EFFL, M10000580 lens) M0161 with new lens (narrow) : MDK-M0161-1-08 (3.24 EFFL, M10001028 lens) It seems that there maybe some discrepancy in the manufacturer stated EFFL for the new lens (3.24 vs original 3.1) because the actual FOV change is more significant than the change in the effective focal length would result in. We are looking into this. Starling 2 and Starling 2 Max drones are currently shipping with the new lens (narrow FOV). We do currently have stock of the old (wide) lenses that should have no issues (according to the manufacturer), however we cannot provide any guarantees for this lens due to past history of issues. If you would like to update your drone / camera with the original wide lens, please contact us and we will make it available for you to purchase (lens part number M10000580). Alex