• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 31 subscribers
  • Views 312 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.

AWR1843BOOST: Visualizer Errors

Ron Caylor
Prodigy 170 points
Part Number: AWR1843BOOST

Tool/software:

I am using the Out of Box demo for the 1843 with the following...

- mmWave_Demo_Visualizer 3.6.0

-  mmwave SDK 3.6.2

- radar toolbox 3.0.0.5

- mmWave Sensing Estimator 2.4.1

I am using the Basic Chirp Creation of the Sensing Estimator to optimize performance for our application.  When attempting to load the generated configuration via the mmWave Demo Visulaizer, the console reports "Error -1" when attempting to execute "sensorStart" and the GUI reports "Error:  Incorrect config reported by target...", even though the mmWave Sensing Estimator did not report any issues with the configurations.

How do I get a better error to understand what is wrong with the configuration?  Since I select the appropriate device and EVM board, why isn't the estimator properly evaluating the feasibility of the chirp design?

Thank you,

Ron

  • 0
    TI__Mastermind 22605 points

    Hi,

    Thanks for posting. Could you let us know what market and end application you are targeting?

    Thanks again,

    Ajay

  • 0 in reply to Ajay Narayan
    TI__Guru*** 148804 points

    Hi,

    As mentioned in the Sensing Estimator, not all chirp/frame configuration generated by the Sensing Estimator will be supported by the SDK demo.

    The reason for that is memory organization in the SDK demo. 

    So, usually we recommend to start with an existing configuration and modify it to check if it works. 

    Thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 170 points

    So, there is no feedback as to why a particular configuration didn't work?  We just have to do trial and error?

  • 0 in reply to Ron Caylor
    TI__Guru*** 148804 points

    The demo at run time will provide some error information that is sent to the visualizer through the UART. This is the message you see in the Visualizer.

    For more information, one should connect CCS debugger and check where the demo fails.

    In this case, it seems that the issue is not the memory but some RF configuration parameters are not correct. 

    "Error -1" when attempting to execute "sensorStart" and the GUI reports "Error:  Incorrect config reported by target..."

    What I would usually do in this case is compare the config generated by Sensing Estimator with one of the configs provided in mmWave SDK OOB demo.

    Thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 170 points

    For our application, we are attempting to detect small drones (roughly the size of a baseball) at maximum distance (50m+).  We are only concerned about 3D position, not velocity nor range resolution.

    Trial and error with the wwWave Sensing Estimator is very frustrating.  Even the "Short Range Default" for the xWR1843BOOST is not accepted by the Out of Box demo.

    Any guidance you can provide here would be much appreciated!

  • 0 in reply to Ron Caylor
    TI__Guru*** 148804 points

    Hi,

    With this demo you will not be able to achieve this performance.  With the AWR1843BOOST we can only detect moving cars with MIMO processing chain at 50m.

    We don't think that you will be able to detect the drones with this radar sensor

    thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 170 points

    Hi Ceasar,

    Note: In my previous posted I wrote "50m+".  That should have read "up to 50m".

    Perhaps you can help reconcile a discrepancy we are seeing.  Our customer has used an IWR1853BOOST and flashed it with the prebuilt MRR demo binary.  With it, he was able to reliably track a drone out to 40 to 50 meters (depending upon how well he stayed in the radar view).  Our radar guy believes that the rotating propellers on the drone enhance its RCS and could explain the performance our customer is seeing.  I am not having similar results with the OOB demo.  (We would have stuck with the MRR demo, but was told that it does not support both azimuth and elevation...even though documentation for the demo states otherwise).  Is the difference in performance between MRR and OOB demos possibly explained by the USRR MIMO processing of the MMR demo?  Is there any other demo similar to the MRR demo (but not so old), that provides similar MIMO processing but with both azimuth and elevation?

    Thank you again,

    Ron

  • 0 in reply to Ron Caylor
    TI__Expert 3330 points

    Hello,

    The USRR subframe is meant for ultra short ranges (to about 20m) so I would attribute the detection at 40 to 50m to the MRR subframe. Unfortunately, not there is no new demo in production that provides the specifications you are looking for. But, to give you something to try, try using the MRR configurations present in the following folder

    C:\Repo\radar_toolbox\source\ti\examples\Automotive_ADAS_and_Parking\medium_range_radar\src\1843\common

    They are not .cfg files so you will need to port the values to a configuration.

    Best,

    Gabriel