• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 632 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

CCS/IWR6843: IWR6843 3D people counter output file

gil albo
Intellectual 725 points
Part Number: IWR6843

Tool/software: Code Composer Studio

Hi,

I'm using the 3d people counter on the IWR6843  board using the updated lab GUI (for 3d ) and cfg file.

The radar is mounted on 1.8[m] tripod with ~10 degrees tilt in elevation ,towards the ground (as appears in the instructions)

The GUI stores the data sent through the UART.

each row of the stored data contains the following information:

target list, height and more.

target list contains Cartesian coordinates , velocity per axis, Dimension per axis. all per detected target.

  1. X
  2. Y
  3. Z
  4. Velocity X
  5. V Y
  6. V Z
  7. Dimension X
  8. D Y
  9. D Z

1. Can you explain what is the difference between Z to height ?

2.How height is calculated.

3.I've recorded a person walking at this configuration

   Radar is placed at (x=0,y=0)

   person starting point (x=2,y=4)

   The person walks at angle of 45 degrees towards the radar at around 0.3[m/sec].

   The height of the person is 1.68[m]

   The used cfg file is attached.

   The fHist.mat is attached

   When examining the height and z values (from the matlab gui output file fHist.mat) along the detected targets, the values aren't stable.

   Here i show the following:

  •   Top view of targets along the walk
  •   target height column values
  •   target z values
  •   target volume =Dx*Dy*Dz

Please explain why the the z,height ,and volume are not stable along the target's track.

Thank you

Gil

2063.mmw_pplcount_demo_default2.cfgshort_man_walk.zip

  • Former Member
    0 Former Member
    Hello Gil,

    The points returned by the tracker are not transformed or correct to account for the sensor's mounting position. The sensor does not know the height or downtilt you have applied.

    Therefore there are two issues:
    1) The target dimensions and position need to be translated to your spatial plane.
    2) There will be variability in the point cloud as you move if you are using the ISK sensor since it has limited elevation FOV. This means that you could be getting reflections from a different cross section of the person as they approach the sensor. This would make the point cloud for the person shift and the resulting dimensions would change. A wider FoV antenna would be less likely to have this issue.

    Amanda