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.

IWR6843ISK: Long range people detection: incompatible with people counting visualizer

Part Number: IWR6843ISK

Hello everyone,

In the forum 'https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1013268/iwr6843isk-long-range-people-detection' it was mentioned that people counting visualizer can be used long-range people detection and I had used that. But now when I was using I got an error "Error: Invalid usage of the CLI command\n". Now it is not working.  I want to use that visualizer.

Thanks in advance.

Thank and regards, 

Aleena N A

  • Hello Aleena,

    It seems that the config file for LRPD in the visualizer folder is outdated. Please make sure you use the config file provided in the LRPD lab folder. Sorry these files got out of sync. I just tried with the 50m 3D profile from the LRPD lab and it worked. However, this visualizer is not optimized for long range so the graph looks a little strange with the longer draw distances. It should work well though.

    C:\ti\mmwave_industrial_toolbox_4_9_0\labs\long_range_people_detection\68xx_long_range_people_det\chirp_configs

    Regards,

    Jackson

  • Hello Jackson,

    Thank you so much. It worked for me.

    Thanks and regards,

    Aleena N A

  • Hello Jackson,

    Is it possible to do long-range people detection visualizer using python? 

    Thanks and regards,

    Aleena N A

  • Hi Aleena,

    What do you mean 'using python'? I thought you were already using the python based visualizer? The python based visualizer (usually used for regular 3D people counting) does also support the LRPD binary/project.

    Regards,

    Jackson

  • Hi Jackson,

    Yes, I was using the python-based visualizer. But as you said the visualizing is strange. Not sure the detection will be accurate for multiple people. In the Matlab visualizer, the detection is almost accurate.

    Thanks and regards,

    Aleena N A

  • Aleena,

    The visualization should still be accurate, it is just a little more difficult to orient the points visually with the 3D grid. The tilt angle of the sensor isn't used, but this will make little difference at longer ranges.  You could rotate the grid to match the 2D view as in the matlab visualizer. Also, you may want to modify the grid draw distance to match the boundary box. Right now it is set for shorter range.

    Regards,

    Jackson