• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 30 subscribers
  • Views 532 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.

IWR6843AOPEVM: Long range small object detction

D J
Intellectual 300 points
Part Number: IWR6843AOPEVM
Other Parts Discussed in Thread: MMWAVEICBOOST

Tool/software:

Hello Ti Expert,

I want Long range small object detection like screws and nuts at approx. 55-60meter range. Angle would be 60-65 degrees.
Can you guide me how to solve this case?

I have also MMWAVEICBOOST for this module.

Thanks in advance. 

  • 0
    TI__Mastermind 23177 points

    Hello,

    You will not be able to detect screws and nuts at those ranges. You can check this antenna overview page for approximate max ranges for each EVM for a person:

    Antennas Overview (ti.com)

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Intellectual 300 points

    Hi Pedrhom,

    Thanks for the response. 

    1) Is it not also possible after the reducing the Azimuth FOV angle and Elevation FOV angle?
    2) Is there any other mmWave hardware, which can achieve this goal?

    Thanks...

  • 0 in reply to D J
    TI__Mastermind 23177 points

    Hello,

    I do not think any millimeter technology in general would be able to detect screws and nuts at those distances even in boresight. In theory even if you could detect it, you would need crazy high resolution to differentiate it from the floor or whatever is around it.

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Intellectual 300 points

    Thanks Pedrohom.

    Now we had a discussion with our end customer and they agreed to see the demo for the people counting application. We have checked the sample code for the people counting but it shows 6m maximum distance. Does it able to show the distance as mention in the Antenna overview document which you have mention above. This we are going to test in the open field area to detect Human or Animal. Can we detect 2D / 3D approximately 90 to 100 meters.

    What are key changes we should do in the program which is available or do you have any such sample program can you share it. Do we test with the AOP only or we need to attached booster as well. Please guide as we need to revert and meet our customer with the solution on coming week Tuesday. 

  • 0 in reply to D J
    TI__Mastermind 23177 points

    Hello,

    As shown with the prior linked antennas page, the absolute maximum for human detection with 6843AOP is about 60 meters. This is the maximum distance in which a person would have a high enough signal-to-noise ratio (SNR) to create a single point which differentiates it from its surroundings and noise. People tracking requires enough detections/points to create a cluster and then on chip algorithms are used to track that cluster's position velocity and acceleration. You will have shorter ranges with people tracking as opposed to people detecting, but you can tune configuration files using the mmWaveSensing estimator's advanced tab where you can copy-paste a cfg's contents into the text box, hit auto populate, and see that config's resulting radar performance. Then you can change parameters which will affect and increase max range.

    https://dev.ti.com/gallery/view/mmwave/mmWaveSensingEstimator/ver/2.4.0/

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Intellectual 300 points

    Hi 

    I created this parameters using mmwave sensing estimator. can you validated it?

    profileCfg 0 60 2 6.4 745.73 0 0 0.34 1 2215 3000 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 3 0 10 1 0

    Thanks,,

  • 0 in reply to D J
    TI__Mastermind 23177 points

    Hello,

    That almost certainly will not work. The sensing estimator is not perfect and does not have catch all for every possible issue, but that Ramp End time of 745.73 with a periodicity of 10 will not work from a timings perspective. Frame periodicity should be generally no faster than 50-100ms.

    If you want very long ranges, I would get one of the configurations from the Long Range People Detection example demo to start off with as a baseline.

    Best Regards,

    Pedrhom

  • 0 in reply to Pedrhom Nafisi
    Intellectual 300 points

    Hello,


    As you mention above, we would to request to provide configuration form the long range people detection example. we can start.




    Above highlight our device. We have also booster device.

    we want to achieve the this range mention in screenshot or higher basis on booster available.


    Thank you.

  • 0 in reply to D J
    TI__Mastermind 23177 points

    Hello,

    You can achieve those ranges for detecting humans. Start with one of the configurations found within the Radar Toolbox within the directory of radar_toolbox\source\ti\examples\Long_Range_People_Detection\chirp_configs. You can download the Radar Toolbox from the link I had earlier with the Antenna Overview page, as well as here https://dev.ti.com/tirex/global?id=RADAR_TOOLBOX

    Best Regards,

    Pedrhom