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-ODS: 3D People Counting

Part Number: IWR6843ISK-ODS

Hello! 

I am setting up the IWR6843ISK-ODS radar sensor for a ceiling mounted 3D people counting application and am running into some issues with the tracking layer parameters.

The application is a small entry room with two doors (entrance/exit). I am attempting to count the number of people entering and exiting the room, I have the parameters for the tracking layer updated for the staticBoundryBox, boundryBox, and presenceBoundryBox, but am having issues with the program not counting the correct number of people entering and exiting. 

The point clouds for individual people are close to each other and moving with the same speed and direction and the radar is detecting them as one person instead of two.

As per the 3D_people_counting_tracker_layer_tuning_guide (1).pdf tracker layer tuning guide it mentions in 4.3 changing the gatingParam and allocationParam to fix the issue of too few people being counted. Is there a recommendation for the best configuration for these parameters? 

Below is an example of the code I am running. Any help is appreciated, thank you!

% SDK Parameters
% See the SDK user's guide for more information
% "C:\ti\mmwave_sdk_[VER]\docs\mmwave_sdk_user_guide.pdf"
sensorStop
flushCfg
dfeDataOutputMode 1
channelCfg 15 7 0
adcCfg 2 1
adcbufCfg -1 0 1 1 1
lowPower 0 0
bpmCfg -1 1 1 2

% Detection Layer Parameters
% See the Detection Layer Tuning Guide for more information
% "C:\ti\mmwave_industrial_toolbox_[VER]\labs\people_counting\docs\3D_people_counting_detection_layer_tuning_guide.pdf"
profileCfg 0 60.75 30.00 25.00 59.10 0 0 54.71 1 96 2950.00 2 1 36
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 6
chirpCfg 2 2 0 0 0 0 0 6
frameCfg 0 2  48  0 55.00 1 0
dynamicRACfarCfg -1 4 4 2 2 8 12 4 12 5.00 8.00 0.40 1 1
staticRACfarCfg -1 6 2 2 2 8 8 6 4 8.00 15.00 0.30 0 0
dynamicRangeAngleCfg -1 0.75 0.0010 1 0
dynamic2DAngleCfg -1 3.0 0.0300 1 0 1 0.30 0.85 8.00
staticRangeAngleCfg -1 0 8 8
fineMotionCfg -1 1 1.0 10 2
antGeometry0 -1 -1 0 0 -3 -3 -2 -2 -1 -1 0 0
antGeometry1 -1 0 -1 0 -3 -2 -3 -2 -3 -2 -3 -2
antPhaseRot 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1
fovCfg -1 70.0 70.0
compRangeBiasAndRxChanPhase 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0

% Tracker Layer Parameters
% See the Tracking Layer Tuning Guide for more information
% "C:\ti\mmwave_industrial_toolbox_[VER]\labs\people_counting\docs\3D_people_counting_tracker_layer_tuning_guide.pdf"
%This is relevant to cars coordinate system  -y y x -x -z z
staticBoundaryBox -0.9 0.9 0 1 -0.5 1.5
boundaryBox -5 5 0 3 -0.5 3
sensorPosition 1.5 0 90
gatingParam 3 2 2 3 4
stateParam 3 3 12 500 5 600
allocationParam 20 100 0.1 5 0.1 0.6
maxAcceleration 0.1 0.1 0.1
trackingCfg 1 2 800 30 46 96 55
presenceBoundaryBox -2.0 2.0 0 1.5 -0.5 1.5
sensorStart
  • Hi,

    The suggestion would be to tune for the use-case. Reducing these parameters should make it easier to detect more people.

    Best,

    Nate

  • Do you have any suggestions for specific parameter values? I have tried the default parameters for ceiling mounted and reducing the parameters suggested in 4.3 by 25% 50% along with a custom parameter set; gatingParam: 3 2 2 3 4 allocationParam: 20 100 0.1 5 0.1 0.6

    I also have used a few of the different chirp configs for 3D_People_Counting and overhead_3d_people_counting and am still having issues where if two targets are closely behind each other moving at the same speed and direction it detects them as one instead of two separate targets. 

    Any suggestion or feedback would be greatly appreciated, thank you!

  • Hi Calvin,

    The procedure you're describing is what we would do. In general, I would recommend modifying the parameters individually instead of all of them as a set. Reducing the number of points needed for a track and shrinking the radius in the gatingparam would be the procedure I would suggest. In general, you are working with one of the more challenging use cases for radar. I would suggest trying a couple different configurations and performing a few trials for each to see what is the best overall.

    Best,

    Nate