• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 1505 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: Changing FPS rate of the radar sensor

Javid Aslan
Prodigy 40 points
Part Number: AWR1843BOOST

Hello,

I am taking Traffic Monitoring experiments together with radar and camera hardware, and using GUI provided in Traffic Monitoring lab.

Frame rate of the camera is 15 FPS and duration of one of the measurements is 9.56 minutes (596 seconds). I checked Radar data and it is composed of 4142 frames which lasted for 2.58 minutes (178 seconds) on Play Back mode. According to that, I found radar's frame rate is around 23 FPS.

Since I am planning to stream video and radar data together and in synchronization I need to make their frame rate equal. I assume the only possible way of doing it is changing some relevant commands in MATLAB script of the GUI. I wonder if I can get some explanation on how to change this rate and which parameters must be adjusted in the code in order to have the same frame rate with visual data.

Thanks!

  • 0
    TI__Guru*** 148514 points

    HI,

    The frame period (ms) is defined in the configuration.

    I am not clear which target code you are using. Is it the mmWave SDK demo?

    Most of our demos are configured using information in a file called "file_name.cfg"

    Here is an example

    mmwave_automotive_toolbox_2_9_1\labs\lab0004_obstacle_detection\chirp_configs\ods_default_config.cfg

    In this file there is a line that defines the frames rate. For example in example below the frame period is 100ms

    frameCfg 0 1 32 0 100 1 0

    Thank you

    Cesar

  • 0 in reply to Cesar
    Prodigy 40 points

    Thank you.

    In my configuration file Frame duration is 66.667 msec and frameCfg and trackingCfg parameters are described below:

    frameCfg 0 1 128 0 66.667 1 0

    trackingCfg 1 1 250 15 230 370 66.667

    Although this frame time corresponds to 15 FPS, when I play gathered radar data from 10 minutes of measurement, it lasts only for around 3 minutes. In parallel, I stream video data from the same experiment (camera's frame rate is 15 FPS) as well.

    So, in simple words, for 10 minutes of real experiment, I have only 3 minutes of radar data. It can be seen that radar worked and recorded all the process during this 10 minutes but in play back mode this data was squeezed into 3 minutes (almost 3x faster than real time). 

    Also, as I mentioned above, frame rate is specified as 15 FPS in frameCfg parameter but, 

    Javid Aslan said:
    I checked Radar data and it is composed of 4142 frames which lasted for 2.58 minutes (178 seconds) on Play Back mode. According to that, I found radar's frame rate is around 23 FPS.

    seems radar worked (or it's just for Play Back mode - unsure about it) with 23 FPS rate.

    I wonder what is the reason of that and how it can be solved - to be able to use Play Back mode in normal speed.

  • 0 in reply to Javid Aslan
    TI__Guru*** 148514 points

    Hi,

    What demo are you using?

    thank you
    Cesar

  • 0 in reply to Cesar
    Prodigy 40 points

    Hi,

    Industrial Toolbox 4.5.1 > labs > Traffic Monitoring > 18xx 68xx Traffic Monitoring > GUI > tm_visualizer

  • 0 in reply to Javid Aslan
    TI__Mastermind 23725 points

    Hello

    The algorithm chain, frame rate and sensing,chirp parameters are tuned in tandem and changing  the frame rate may have adverse effect on the  performance and timing of the demo.

    It is doable and could entail balancing the rest of the processing elements along with it.

    Would you have the option of using only the required frames from the mMwave data set as this would be least invasive of the routes to the synchronization going between two sources.

    Thank you
    Vaibhav