• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 437 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.

IWR6843AOP: Target Detection with Speeds Above Max Velocity

Spark
Prodigy 80 points
Part Number: IWR6843AOP

Tool/software:

Hi there,

I’m working on a modified version of the Area Scanner demo, using the following profile configuration:

profileCfg 0 60 100 10 79 0 0 24 1 256 4000 0 0 30  
channelCfg 15 7 0  
adcbufCfg -1 0 1 1 1  
chirpCfg 0 0 0 0 0 0 0 1  
chirpCfg 1 1 0 0 0 0 0 2  
chirpCfg 2 2 0 0 0 0 0 4  
frameCfg 0 2 32 0 70 1 0


This setup achieves a max range of 20 m and a range resolution of 0.098 m, which is ideal for my application. However, the max velocity is relatively low at 2.288 m/s, and I’m unable to make significant changes to the configuration without compromising other requirements.
My question is: What happens if a target exceeds the max velocity (e.g., a target moving faster than 2.288 m/s)?
From my understanding, the radar will still detect the target, but the velocity will be ambiguous or incorrect, while range and angle measurements should remain unaffected. Is this correct?
I’d appreciate your insights or confirmation.

Thank you for your support!

Best regards,
Ed

  • 0
    TI__Mastermind 22577 points

    Hello Ed,

    It is as you said, it is not that the object will suddenly disappear, but rather be smeared across several doppler bins and potentially looped back to "0". You can think of it being somewhat similar to when a picture is taken of a very fast object with the camera configuration not being set for fast objects, the resulting image shows the fast object smeared. The individual points in the point cloud will be accurate on a range perspective, but doppler will essentially "overflow" and give an inaccurate and incorrect measurements. There are certain ways of allowing for extra velocity to be showcased, and there's some information here:

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1000895/iwr6843isk-estimate-velocity-beyond-the-max-velocity-in-long-range-people-detect-lab

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Prodigy 80 points

    Hello Pedrhom,

    Thank you for the clarification, it’s reassuring to know that even if the velocity becomes ambiguous, the range and angle measurements will remain accurate. That’s an important confirmation for my application.

    I have a couple of follow-up questions:

    1. SNR and Noise in Side_Info:
      Will the Signal-to-Noise Ratio (SNR) and noise values in the side_info remain unaffected when the velocity exceeds the maximum? Or are there any potential distortions in these metrics as well?
    2. Multi-Mode for Higher Velocity:
      What you’ve proposed is very interesting. If I understand correctly, this would involve operating the radar in multi-mode with multiple chirp profiles to accommodate higher velocities. Does the Area Scanner demo support such a multi-mode operation, or would this require significant modifications?
    3. extendedMaxVelocity Command:
      I’ve noticed the extendedMaxVelocity command, which seems designed to increase the maximum velocity, but it’s not supported on the IWR6843AOP. Do you know why this limitation exists? Is it related to the hardware or firmware constraints of the AOP version?

    Thank you again for your insights and support!

    Best regards,
    Ed

  • 0 in reply to Spark
    TI__Mastermind 22577 points

    Hello Ed,

    SNR and CFAR calculation accuracy are independent of doppler value. There are however parameters within the configuration of CFAR that will exclude points based on their doppler value, so this could be something to look at for.

    As far as multi-mode for higher velocities, it is possible that the pointcloud generation will still work as the Area Scanner demo is based on the Out of Box demo, but I would say it is almost certain that the heatmap generation and static point plotting will not work as intended. You can find an example of advanced subframes and multiple profileCfgs (chirp designs) within the mmWave SDK at C:\ti\mmwave_sdk_03_06_00_00-LTS\packages\ti\demo\xwr68xx\mmw\profiles. These configurations are meant to be run with the out of box demo binary, which for the 6843AOP is xwr64xxAOP_mmw_demo (this is not a typo).

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Prodigy 80 points

    Hello Pedrhom,

    Thank you for the clarification, all clear.
    If I can inquire just about one last option, I've noticed that another command exists (extendedMaxVelocity) that enables the detection of velocities up to two times Vmax. But on the mmwave_sdk_user_guide it's written that this command is not supported on IWR6843AOP. Do you perhaps know if this is a hardware or firmware limitation?

    Thank you again for your insights and support!

    Best regards,
    Ed

  • +1 in reply to Spark
    TI__Mastermind 22577 points

    Hello Ed,

    It seems to be not supported with specifically AOP EVMs. I looked at the DFP firmware and SDK function calls and there is nothing stopping a user from enabling this parameter in the 64xx_aop (6843AOP) binary, but it seemingly just does not do anything as there are no crashes. So my answer is that in theory one could implement an extended max unambigious velocity algorithm for the AOP devices, but we ourselves do not have one.

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Prodigy 80 points

    Hello Pedrhom,

    Thank you, it's good to know that there's no hardware limitation.

    Best regards,
    Ed