• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 385 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.

IWR1443BOOST: Sensor Tuning

Christoph Hintz
Prodigy 125 points
Part Number: IWR1443BOOST

Hi eerybody,

I wanted to ask about sensor tuning for the mmwave radar. Is there a guide on tuning the sensor. Looking at the configuration file it is not visible what we have to change to get certain results. We have the AWR1433Boost with a configuration that sees many obstacles on our robot. As an example if I walk in front of the robot it gives me like 10 points, which seems good, however if I roll a soccer sized ball on the ground there is no feedback. I just wanted to check if there is a guide with explanations or maybe somebody we could contact to get some better background on the tuning parameters.

Best,

Christoph

  • 0
    TI__Mastermind 30405 points

    Hi Christoph,

    Please refer to the following thread for a general discussion on the various factors that determine the design of sensor configuration parameters for FMCW radar in general and TI mmWave Sensors.

    IWR1443: Generic development flow to define/implement the RF and DFE configuration for the specific application

    You've not mentioned whether you're using the SDK Out-of-box demo or your own software. If you're using the OOB demo software, you have a few options to tune the sensor for your detection needs.

    1. Selecting the right chirp parameters depending upon the RCS of your object (the smallest one you want to detect), max range and max velocity is the first step. The OOB demo visualizer provides preset configurations depending upon your most important requirements (best range, best range resolution or best velocity).

    2. Try to detect the object at rest first and see if it shows up (as a peak) in the 1D FFT plot for Zero Doppler on the visualizer. If you cannot see the object at this stage, you need to see if the Object is reflective enough (shape, material, in essence RCS of the object) and play with the Chirp parameters more. Again, use the Sensing Estimator.

    3. Enable the Range-Azimuth Heatmap also to get a better picture of the scene before it goes through the CFAR detection algorithm (where peaks are filtered out based on the CFAR algorithm).

    4. If you can see the object in the Range-Azimuth heatmap but it doesn't show up in the X-Y scatter plot, then you can adjust the CFAR detection threshold to a lower value (say 10 dB instead of the 15dB default) to see more points.

    5. You can also disable peak-grouping to see more detected points on the scatter plot. Peak grouping is enabled by default and it reduces the number of points sent out by the software by grouping neighboring points into one.

    Please mark this thread resolved if your question was answered otherwise get back if more support is needed.

    Regards

    -Nitin