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-ODS: Obstacle detection demo gives bad results

Part Number: AWR1642BOOST-ODS
Other Parts Discussed in Thread: AWR1642BOOST

I am using the Obstacle Detection Demo with the Matlab exe file (Here is the link:  http://dev.ti.com/tirex/#/?link=Software%2FmmWave%20Sensors%2FAutomotive%20Toolbox%2FLabs%2FObstacle%20Detection). The config file is also provided within the demo. But I get very strange results. Detected objects are in the air, it looks very noisy and it also detects very often the wall or the table as a cluster. I reduced the min. nr. needed for the cluster, which resulted in slighty better results. I only want it to detect humans. The keyword clutterRemoval is also not recognized by the board, so static object also get clustered sometimes. Could anybody help me or give me a good config ?

  • Hi,

    The demo was designed to detect obstacles. So, it will detect objects in the field of view. It is not possible to detect only humans.

    So, it will detect objects such as tables or walls.

    You may want to consider other processing chains such as People Counting Demo

    thank you
    Cesar
  • Hi,

    after your suggestions I switched to People Counting Demo. It shows the same strange results for my ODS-Board. In the near left part of the sensor it detects me, in front of the sensor and right it does not detect me anymore. Than I used the AWR1642BOOST without ODS, and this works perfectly fine. Do I have to change sth additionally in the configs when I use the ODS?

    thank you

  • Hi,

    ODS and BOOST EVMs have different antennae.

    This means that they required different code implementations

    A demo can run correctly either on ODS EVM or on BOOST EVM but not on both.

    So, the People Counting Demo can only run on BOOST EVM and not on ODS

    thank you
    Cesar