• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 1304 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.

AWR1642BOOST: Finding angle of metal beam

Joakim Hesselgren
Prodigy 20 points
Part Number: AWR1642BOOST

Hi, 

I am looking into to see if the AWR family is possible to be used to find the angle of a metal beam that is moving towards the radar (picture added for more description). The closest point the beam would come to the radar would be 15-20 cm. Would this be possible to be be done with the AWR family and which one would be prefered? Is it possible to say how good acurracy the angle could be at the closest point?

Have been testing the AWR1642BOOST so far but only with the demo visualizer with some trouble of finding the beam at some points and during those times it only shows 1 point in the GUI.

Have not found exactly what the demo visualizer nor the DSP does with the data so I am wondering also if by using the raw ADC data or the data sent over UART it would be possible to see the area of the beam or with multiple points rather than just 1 to be able to find the angle of it? 

  • 0
    TI__Mastermind 30405 points
    Hi Joakim,

    Looking at the diagram, it should be possible to determine the angle of the beam w.r.t. the sensor. A few points to consider:

    1. What is the accuracy you are looking for? Additionally, it is important to differentiate between Accuracy and Resolution. Please look at the following thread which explains these and other important concepts related to angle estimation.

    e2e.ti.com/.../2212393

    2. Depending upon the size and orientation of the beam w.r.t the sensor and depending upon the chirp configuration, it may span multiple range bins and so you may get more than one range peaks in the Range-FFT. Similarly, it may span multiple angle bins as well so your application code would need to be able to single out the range/angle of interest (you can look at a particular range bin, e.g. the nearest part of the beam from the sensor and get the angle for that).

    3. You can look at the visualizer source as explained in the following thread:

    e2e.ti.com/.../610099

    Regards
    -Nitin
  • 0 in reply to Nitin Sakhuja
    Prodigy 20 points
    Hi again,

    The accuracy that is being desired would be about 1 degree and atleast below 5 degrees. With accuracy I mean how much can the measured angle of the beam differentiate from the real value?

    So it is possible to look at a small part of the beam (lets say 5x5 cm large) and by that get the angle of the beam?

    Not sure how I missed the first couple of times of using the demo visualizer but noticed there was a group option which fixed some of my concerns. Been testing on a small metal plate at the moment and it can find it easily when it is parallell with the radar but when increasing the angle by a small amount it disapears. My guess is that there comes no reflections back to the radar so I would assume that I would need some kind of reflector on the beam?

    Regards
    Joakim
  • 0 in reply to Joakim Hesselgren
    TI__Mastermind 30405 points

    Hi Joakim,

    Sorry for the delay in getting back. Your second observation regarding difficulty detecting the metal plate as the angle increases seems plausible because increasing angle would cause the signal to be deflected resulting in less signal being reflected towards the sensor. Putting a corner reflector would help help reflect more energy to the sensor and hence improve performance.

    Please refer to the System Performance App note given below to get information about achievable angular accuracy.

    www.ti.com/.../swra549.pdf

    Regards
    -Nitin

  • 0 in reply to Nitin Sakhuja
    TI__Mastermind 30405 points
    Hi Joakim,

    Do you have further questions on this subject? In case the above responses answered your question, can you please close this thread?

    Thanks
    -Nitin
  • 0 in reply to Nitin Sakhuja
    TI__Mastermind 30405 points
    Closing this ticket for lack of activity. Please create a new ticket if you have further questions. Thanks.