• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 274 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.

IWR6843ISK: Evaluation Board Suitability for Vehicle Height Profiling

Florian Baumgartner
Prodigy 20 points
Part Number: IWR6843ISK

Dear TI Support Team,

We are currently developing an application aimed at measuring the height profiles of vehicles using mmWave radar technology. For this, we have employed the IWR6843ISK evaluation board, utilizing the standard "Out of the Box" demo firmware.

Our setup involves mounting the sensor approximately 4.0 meters above the ground, oriented downwards. Despite this, our trials have been somewhat disappointing. The resulting point-cloud data often includes numerous erroneous or irrelevant points, and on occasion, it fails to detect highly reflective objects altogether.

The scope of our measurements spans a distance of 0 to 10 meters, with a field of view ideally covering both 120° and 30° angles. However, we are particularly concerned about the precision of the angular resolution in elevation, which is supposed to be ± 15°.

Given these challenges, we seek your expertise to confirm whether the IWR6843ISK sensor is appropriate for our requirements. Moreover, we would appreciate insights into whether the antenna configuration on this evaluation board is capable of accurately tracking moving vehicles with fine resolution. So that a absolute error of less than 10 cm can be achieved (in the 3D space).

Thank you for your assistance. We look forward to your expert guidance.

Best regards,

Florian

  • 0
    TI__Expert 8970 points

    Hi

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

    Regards

  • 0 in reply to Abhishek Karkisaval
    TI__Mastermind 32830 points

    Hi Florian,

    Height measurement is possible with the mmwave sensor. My recommendation to maximize your performance would be to attempt to orient the range dimension (which is perpendicular to the surface of the PCB) as closely with the height dimension as you can. As you're referencing, the angular resolution and angular accuracy are less than the range resolution and range accuracy.

    Florian Baumgartner said:
    Given these challenges, we seek your expertise to confirm whether the IWR6843ISK sensor is appropriate for our requirements. Moreover, we would appreciate insights into whether the antenna configuration on this evaluation board is capable of accurately tracking moving vehicles with fine resolution. So that a absolute error of less than 10 cm can be achieved (in the 3D space).

    I think the IWR6843ISK is appropriate for this application, but you could also consider the ODS or AOP variants too if you want to balance azimuth and elevation accuracy. 

    As for the absolute error - can you show a picture of the orientation of the board with respect to the cars? And are the cars moving in this scenario? Or are they parked? 

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 20 points

    Hi Nathan

    Thank you for your comprehensive answer.
    We have mounted the sensor facing downwards, so the range dimension should be the same as the height above the ground (4m in our example).
    Please have a look at the attached photos of the measurement setup.

    We will also consider using the ODS board for further tests, as we definitely need both high azimuth and elevation resolution.
    The cars are moving. In this first example only in walking speed. But later the goal is to capture moving vehicles with a framerate up top 100 Hz.

    How would you proceed on conducting further measurements? Which of your (TI) examples cover our application the most?


    Thank you and best regards

    Florian

  • 0 in reply to Florian Baumgartner
    TI__Mastermind 32830 points

    Hi Florian. I don't see any pictures attached. Please try again.

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 20 points

    Ok, sure I can post them again. Have a look here

  • 0 in reply to Florian Baumgartner
    TI__Mastermind 32830 points

    Thank you Florian,

    I think this setup should be accurate for estimating height. You will have to verify the accuracy and precision yourself, but I think this is very doable. You could use the small obstacle detection example as a starting point, or just the out-of-box example. I would use a fast frame rate and do some pooling of the data over frames to get a maximum estimate.

    100 Hz may not be feasible for the radar sensor, but I think 20-30 Hz would be. Is that acceptable?

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 20 points

    Hi Nathan

    Thanks tor the clarification. We will switch now to the ODS Antenna configuration as it seems better suited for our purpose.
    What is the limiting factor of the frame rate (chirp rate, processing time, communication)?
    What would you suggest to focus the most on if we would like to achieve a high frame rate?

    Best,
    Florian

  • 0 in reply to Florian Baumgartner
    TI__Mastermind 32830 points

    Hi Florian,

    Processing time will most strongly dictate the maximum frame rate you can achieve.

    Best,

    Nate