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 counting Lab

Part Number: IWR6843ISK

Hello there,

Sensor Board = IWR6843ISK

Industrial toolbox = 4.7

Lab = Long Range People Detection

Chirp Config = people_detection_and_tracking_100m_2D_advanced.cfg

1. I am working on Long Range People counting Lab (with 100m 2D chirps-configs) provided in the "industrial toolbox 4.7". And, I am not able to run this lab in traffic monitoring GUI with the above Chirp-Config, as described in the docs. How can I run the above lab with the above chirp configuration in traffic monitoring GUI?

It works perfectly with chirp Config-"people_detection_and_tracking_50m_2D.cfg".

2. However, it does run on People Counting (python) visualizer, but the performance on that visualizer is the worst, it never track people properly when I use above lab & chirp config.

3. I am using this sensor directly with my PC (without EVM board) , but in docs it is recommended to use 5V-3A power source, do you think this could be the reason for poor performance? 

    Is there any way to provide external power to the sensor board?

  • ***This is the error ********************

    COM14 opened.
    COM13 opened.
    Opening configuration file
    Parsing configuration file...
    Skip parsing for: sensorStop command
    Skip parsing for: flushCfg command
    Skip parsing for: %compRangeBiasAndRxChanPhase command
    Skip parsing for: %SceneryParam command
    Error: number of parameters is incorrect for stateParam commandError: number of parameters is incorrect for trackingCfg commandSkip parsing for: sensorPosition command
    Skip parsing for: presenceBoundaryBox command
    Skip parsing for: sensorStart command
    Error using *
    Too many input arguments.

    Error in calculateChirpParams (line 116)

    Error in tm_visualizer (line 75)

  • Hello,

    1. There is a bug in the tm_visualizer gui that isn't compatible with the 100m profile. This will be fixed in the upcoming toolbox release. If you have a matlab license, you can modify the source files and run there using the instructions below. Or the demo is also compatible with the people counting visualizer usually used for 3D people counting (not long range) as noted above.
    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/989593/iwr6843isk-100m-cfg-file-in-long-range-people-detection-lab 

    2. Yes this visualizer was not optimized for the really long range profiles.

    3. If you are just using the ISK board, and not the IC Boost board, then regular power from the PC is fine. The higher power supply is only needed when using the IC boost. You can always provide more USB power with a powered USB hub.

    Regards,

    Jackson

  • Thanks Jackson for replying my queries,

    Now tm_visualizer can run 100m profile.

    But there is very low frame rate, is there any way to increase the frame rate in 100m profile?

  • Hello,

    Because the 100m profile uses beamforming, the framerate must be reduced. It is set at 400ms. You may be able to reduce this by modifying the subframecfg commands and the trackingcfg command in the cfg file, but this may cause the subframes to not have enough time to complete.

    Please see the documents at the below path for more info on the beamforming and how it compares to a normal profile. Frame rates settings can be found in the SDK user's guide.

    C:\ti\mmwave_industrial_toolbox_4_7_0\labs\long_range_people_detection\68xx_long_range_people_det\docs

    Regards,

    Jackson