This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

AWR6843AOPEVM: Embedded software - example projects

Part Number: AWR6843AOPEVM
Other Parts Discussed in Thread: AWR6843

Hi,

I own a AWR6843AOPEVM device and would like to run this example or this one.

The thing is that I would like to avoid using a PC for processing. The ideal scenario for me would be to flash an external device with all that's needed for processing, such as Raspberry Pi.

Moreover, in the second example I've mentioned above, it says:

Note: In this version, the AWR6843 outputs the spherical-coordinate point cloud. Zone mapping and decision state machine processing are perform for the short term in Matlab. These will be moved to the xWR6843 in a near term release to the mmWave Automotive Toolbox.

Is it already done? Where can I find that release? 

In the first example I've mentioned above, which seems more up-to-date, such a release isn't mentioned and the use of MATLAB still exists.

If there isn't such a release, I'd like to know if it's possible to use a sort of a compiler to do that myself. Is there a specific one that is recommended? What would be the process of implementing that?

I would appreciate some help with that,

Thanks.

  • Hi,

    Let me check internally if the non-Matlab version exists. I will update shortly.

    Best Regards,
    Kevin 

  • Hi,

    The non-Matlab versions do not exist. I would suggest focusing on the High Accuracy Demo as that has been updated recently. If you download the toolbox, navigate to <toolbox_dir>/tools/visualizers/InCabin_HighAccuracyDemo_GUI/src to find the code for the visualizer. This will include the processing that is done. AWR6843_HighAccuracyDemo_visualizer.m is the main file that processes the TLVs then calls the functions in the other files (e.g.  zone_assign.m, occupancyDetection.m, and classification_logic_perSeat.m). These functions can be ported to your desired processing platform.

    Best Regards,
    Kevin

  • Thank you for checking that and updating me.

    Is there any intention to release a non-Matlab version in the future? I'd appreciate an estimation for that.

    As it was mentioned in the quote I've posted above, it seems that there was such intention in the previous demo.

  • Hi,

    Plans to release a non-Matlab version were deprioritized. This was due to the understanding that users would implement and ultimately optimize their own post-processing chains. The post-processing done in these demos is relatively simple and focus on just being a baseline demonstration of the device's capabilities. Users may develop much more sophisticated layers of processing to achieve desired performance. 

    Best Regards,
    Kevin