• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 144 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

IWRL6432AOPEVM: Inquiry Regarding Detection of Seated Individuals Beyond 3m Distance

SeungCheol Kim
Prodigy 70 points
Part Number: IWRL6432AOPEVM
Other Parts Discussed in Thread: IWRL6432AOP

Tool/software:

Hello,

I am currently testing the following environment:

[RADAR SENSOR]: IWRL6432AOP EVM
[SDK]: MMWAVE_L_SDK_05_05_00_02
[Radar Toolbox]: 2.20.00.05
[Visualizer]: Industrial Visualizer

In the Industrial Visualizer, I detect presence by dividing the area into 4 zones using the following configuration:

mpdBoundaryBox 1 -1.5 1.5 0 1 0 3
mpdBoundaryBox 2 -3 -1.5 0 1 0 3
mpdBoundaryBox 3 1.5 3 0 1 0 3
mpdBoundaryBox 4 -3 3 1 4 0 3
In my test results, the system works well in detecting both standing and seated individuals within a range of 3 meters. However, beyond 3 meters, it detects standing individuals fairly well, but seated individuals are not detected as reliably.

I would like advice on which parameters to adjust in order to improve the detection of seated individuals beyond 3 meters.

The following is the environment I am testing with:

[RADAR SENSOR]: IWRL6432AOP EVM
[SDK]: MMWAVE_L_SDK_05_05_00_02
[Radar Toolbox]: 2.20.00.05
[Visualizer]: Industrial Visualizer
[cfg]
sensorStop 0
channelCfg 7 3 0
chirpComnCfg 15 0 0 128 1 28 2
chirpTimingCfg 6 28 0 75 59.75
frameCfg 2 0 200 64 250 0
guiMonitor 2 3 0 0 0 0 1 0 1 0 1
sigProcChainCfg 32 2 3 2 8 8 1 0.3
majorStateCfg 4 2 20 8 6 40 4 4
minorStateCfg 4 3 10 8 6 20 8 20
cfarCfg 2 8 4 3 0 12.0 0 0.5 0 1 1 1
aoaFovCfg -70 70 -40 40
rangeSelCfg 0.1 5.0
clutterRemoval 1
antGeometryCfg 1 1 1 0 0 1 1 3 1 2 0 3 2.5 2.5
compRangeBiasAndRxChanPhase 0.0 1.00000 0.00000 1.00000 0.00000 1.00000 0.00000 1.00000 0.00000 1.00000 0.00000 1.00000 0.00000
adcDataSource 0 adc_data_0001_CtestAdc6Ant.bin
adcLogging 0
lowPowerCfg 1
factoryCalibCfg 1 0 40 3 0x1ff000
boundaryBox -3 3 0 4.5 -0 3
staticBoundaryBox -3 3 0 4.5 0 3
sensorPosition 0 0 0.9 0 0
gatingParam 6 2.5 2.5 2.5 8
stateParam 3 3 12 50 5 200
allocationParam 3 10 0.1 4 0.5 20
maxAcceleration 1.0 1.0 0.2
trackingCfg 1 2 100 5 61.4 191.8 100
mpdBoundaryBox 1 -1.5 1.5 0 1 0 3
mpdBoundaryBox 2 -3 -1.5 0 1 0 3
mpdBoundaryBox 3 1.5 3 0 1 0 3
mpdBoundaryBox 4 -3 3 1 4 0 3
presenceBoundaryBox -3 3 0 4 0 3
microDopplerCfg 1 0 0.5 0 1 1 12.5 87.5 1
classifierCfg 1 1 4
baudRate 1250000
sensorStart 0 0 0 0
Thank you for your advice.

Sincerely,
SeungCheol Kim

  • 0
    TI__Expert 8990 points

    Hi

    Thanks for your query. Please allow us a couple of days to respond.

    Regards

  • 0 in reply to Abhishek Karkisaval
    Prodigy 70 points

    Hi.

    Thank you for your review.

    Additionally, is there any way to give different values for the following parameters for each mpdBoundaryBox?

    majorStateCfg 4 2 20 8 6 40 4 4
    minorStateCfg 4 3 10 8 6 20 8 20

    Please review this along with the other issues.

    Sincerely,
    SeungCheol Kim

  • 0 in reply to SeungCheol Kim
    TI__Genius 16601 points

    Hi SeungCheol, 

    You will likely need to reduce the minor motion state machine thresholds. The presence detection tuning guide has guidance for the parameters you should tune.

    SeungCheol Kim said:
    Additionally, is there any way to give different values for the following parameters for each mpdBoundaryBox?

    This is not possible with the base motion and presence detection demo without modifying the source code. If you want to try adding this functionality this you can refer to the video doorbell demo source code which implements a version of this allowing for zone specific thresholds for the major motion state machine only.

    Best Regards,

    Josh

  • 0 in reply to Josh Dye
    Prodigy 70 points

    Hello,

    Thank you for your response.

    I have already performed some tuning based on the tuning guide you mentioned, and it has resulted in some improvements.

    With these tunings, the detection of standing individuals has improved to some extent, but the recognition of seated individuals is still lacking.

    I will proceed with further tuning based on the guide as suggested.

    Additionally, I will also review the video doorbell demo source code.

    Sincerely,
    SeungCheol Kim