@Eric-Katzfey interesting, think you for this information. Once i demo this drone for a couple weeks in its current BNF w/ ndaa add ons configuration I will be looking to move to a smart system and incorporating optical hold and other gps denied search features.

@ashley-miller , for this motor (28xx class), you should use the following ESC M0138 FPV ESC :

https://www.modalai.com/products/m0138
https://docs.modalai.com/voxl-fpv-esc-datasheet/

The mini ESC (M0129) is designed for small and lightweight drones and mini FPV racers (750g or less for slow flyers and 500g or less for mini FPV drones).

The peak power for one of these motor is almost 1kW each - the mini ESC cannot handle that for several reasons, such as peak current capacity, thermal.

If you would like to use the 2808 motor, please choose the M0138 FPV ESC and perform motor calibration using voxl-esc tools : https://gitlab.com/voxl-public/voxl-sdk/utilities/voxl-esc/-/blob/master/voxl-esc-tools/calibration.md

Also, please review the following FAQ: https://docs.modalai.com/voxl-escs/faq/

Alex

@jeremyrbrown5 we figured it out, we just changed the name of the model to yolov8n_float16.tflite. Pretty spot on now!

@roy-crosthwaite see here: https://www.modalai.com/pages/rma

Hi @Alex-Kushleyev, thank you kindly for the support and apologies for the late response. My earlier question was for an attempt to equip an M0173 to VOXL2 for a C26/27/28 config, plus an M0010 stereo pair on a M0076 interposer on VOXL J8 (sorry for the lack of clarity). Our team is now looking to use a default camera config for M0173, but minus the ToF sensor and with an added M0161 4K Hires camera connected to VOXL J8 via the M0155 adapter. Planning to test this hardware config shortly

@Jakub-Tvrz said in Starling 2 Max for indoor:

We haven't bought anything yet, but we are deciding what will be the best for our warehouse inventory application. Of course, we considered the ToF option. But you still say: "however, it is not... great" - does that mean that Starling 2 is better for our application? Or does Dual vs Triple camera for Visual Inertial Odometry also have an impact?

Reply

Depends fully on your use case. I will say that PX4's instance of obstacle avoidance is not fantastic, but it is better than nothing. In my opinion I would go w/ the model that has all the bells and whistles in case you need the hardware and decide to develop your own collision avoidance or somethign of that nature.

Hi @dylangresham
Stinger is intended to be a complete drone package for developers, integrators, and system providers who do not require additional customization. We believe it includes all the necessary functionality for the FPV use case.

Starling 2, Starling 2 Max, and the PX4 Developer Editions, on the other hand, are designed to be customizable platforms that developers and integrators can modify to suit their specific products and use cases. By design, these are not fully Blue List–certified turnkey drones, since any additions, removals, or changes made by the end developer would require recertification.

That said, because most (if not all) of the subcomponents used in these platforms are already listed on the Blue UAS components list, the end-user certification process should be straightforward. ModalAI parts will pass screening easily thanks to their existing certifications.

I hope that clarifies the rationale.

@taiwohazeez Can you post a video of it trying to fly?

@jeremyrbrown5 , it is possible that overheating VOXL2 cpu could be causing your SoftAP performance slowdown. You can run voxl-inspect-cpu which is a tool that will show you cpu usage, temps and other stats. If the CPU cores reach 95 deg C, the thermal management will slow down the cpu clock, so all the processes will run slower, including SoftAP.

Alex

@Eric-Katzfey

Thank you for reverting back. I am trying to proceed with the same approach.
I am facing some errors while going with this approach , I am trying to debug them.

Another approach :-
I am also thinking to hardcode the joystick based turtle mode to the voxl_esc.cpp.
The manual control values will be monitored and once the value of the assigned button is clicked, the turtle mode will be enabled.
Do you think it is a good approach?

@lhermosillo
That is harsh. Do you have a bare board, or one with some additions? If you can use the bare voxl2, try that. If you have been using the bare board all along, I'm stumped here. I'd maybe try to reinstall the modalai-qdl env, or use another image. Maybe the qdl command itself instead of the webUI. Hope you'll be able to save your board!

@N-Chen I tried the webUI and the command line and as near as I can tell the webUI uses lsusb (or something along those lines) to detect the presence of the board. The install.sh uses adb or fastboot to detect the board and will sit waiting until the board is detected or CTL C is sent. My guess is the usb controller interface is alive and reporting but everything beyond it is dead.

@Zachary-Lowell-0 Hi, let me explain what i meant by hovering mode. There is a 3-way toggle switch beside the kill switch in the transmitter (toggling down switches to position mode, toggling up switches to the autonomous mode, while toggling to the middle maintains the drone flight positioning. I have another drone that when the button is switched to the middle, it maintains the drone position (thats what i meant by hover mode). My other starling 2 drones maintains their positions when the switch is toggled to the middle position, but this particular starling 2 drone doesnt.

Were you able to solve this issue and unbrick the device? Can you explain how?

You will need the SKU supporting the TOF to accurately run current versions of voxl-mapper and VOA.

Zach

The system has an HDMI out so you can plug it into an external monitor yes:

https://docs.modalai.com/vrx-ground-unit-developer-guide/

@Ido-Goldstein to run yolov8 run voxl-configure-tflite-server and then when it prompts you to select your model, you can select yolov8. YOLOV11 is also currently supported however is still CPU bound so I would recommend using yolov8 for GPU processing

Hi @SKA ,

I believe i fixed this issue with full-follow mode in the camera server. Now EIS has a body->imu transform, which is read from the extrinsics file and is used to properly rotate the image, instead of hi-jacking the cam->cam_eis transform.

So now your cam->cam_eis transform should work. Please make sure the imu_apps->cam transform is correct, if your camera is rotated 90 degrees.

Please check the commits from Nov 10 : https://gitlab.com/voxl-public/voxl-sdk/services/voxl-camera-server/-/commits/dev

Alex