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: Unable to load config file

Part Number: AWR1843BOOST
Other Parts Discussed in Thread: AWR1843

Hi, 

I am using AWR1843BOOST radar EVM. I have tested radar module with "mmWave Demo Visualizer" "dev.ti.com/.../" and it was working fine for detecting object till 20m. Later I have searched on community forum to increase the range I got to know about demo binary. I have flashed demo binary "xwr18xx_mrr_demo.bin" in radar module and I have used "MRRVisualizer" to test. I am able to detect object till 50m range but if object moves in horizontal pane like left or right radar doesn't detect object properly. 

After I tried to load config file from "mmWave Demo Visualizer" but now I am unable to load it in radar module. Radar module is not getting connected with "mmWave Demo Visualizer".

Please let me know about your inputs on this it will be helpful for us.

Thanks,

Manoj

  • Hi,

    Please let us know what application you intend to use the AWR1843 for. This will help us recommend the right demo.

    Please note that the mrr demo binary only works with the MRRVisualizer, it does not work with the mmWave Demo visualizer

    thank you

    Cesar

  • Hi Cesar,

    We want to use this radar in some advanced applications like obstacle detection, it should be able to detect obstacles which are crossing radar in parallel path instead of perpendicular path. We are exploring this radar module and doing some table top activities to check the feasibility.

    We had developed our own serial communication utility to grab the packets received from AWR1843 used  inplace of mmWave Demo visualizer and was working pretty fine. Now with the update of AWR1843 with the demo binary image our utility as well as mmWave Demo visualizer aren't functioning. I guess there is a difference in serial protocol in the demo binary image that is used. Could you share us the protocol details so we can update our utility.

    Since the radar has obstacle detection capabilities up to 150m and using the demo binary image we are able to get a range upto 50m. Could you help us to enable the detection range up to 150m.

    Thanks,

    Manoj

  • Hi,

    The best radar performance can be achieved using the MRR demo. You will not be able to get better performance than beamforming - which enables all Tx.

    You are correct that the MRR demo has a different data format than the mmWave SDK 3.6 Out of Box demo.

    It is important to understand the following information regarding the MRR demo:

    mmwave_automotive_toolbox_3_6_0\labs\lab0007_medium_range_radar\docs\MediumRangeRadar_DeveloperGuide.pdf

    Slide 5 Mode of Operation

    The demo can support two types of frames with 3 different modes of operation. The prebuilt binary is configured for the USRR/MRR alternate subframes.

    If you are only interested in long range, you would need to use the MRR subframe only configuration. See slide 15 on how to do that.

    Slide 10 Output Data Format

    This slide describes the Output Data Format for MRR mode. One can also understand this output data format from the src code of the Matlab GUI (that parses the output format.

    Thank you

    Cesar

  • Hi Cesar,

    Thank you for information. One point that we mentioned earlier is our observation during our testing with MRR demo binary flashed in the radar is that we are able to detect obstacles that are moving perpendicular to the radar(coming closer and going away from the radar), but we are not able to detect obstacles moving horizontally across the radar(moving from left to right and vice versa). 

    Could you help us to enable detection for horizontally moving objects.

    Thank you,

    Manoj

  • Hi,

    Detecting objects moving horizontally across the radar will be more challenging because the radar is only able to detect the radial velocity.

    When an object moves "horizontally", the radial velocity does not change as fast as when the object moves "vertically" This means that when the object moves, it will spend longer time in the same doppler bin. This will make it more difficult to detect. Increasing the Velocity Resolution should improve the detection.

    The Velocity Resolution will improve by increasing the Active Frame Time. This can be increased by increasing the chirp Idle time

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1050220/faq-computing-maximum-range-velocity-and-resolution-mmwave-system

    Of course, the consequence of this will be that the Number of frames per second will decrease because we need to increase the frame period.

    thank you

    Cesar

  • Hi Cesar,

    I have modified the code for long range detection. I have changed it to MRR subframe only configuration. As per developer's guide I am able to rebuild mrr_18xx_dss project successfully but not the mrr_18xx_mss. It takes lot of time and never finishes rebuild process, it stuck at 40%. I tried cleaning up the code before rebuild it, but didn't work. It will be very helpful if you can guide.

    Thanks,
    Manoj

  • Hi,

    Unfortunately we are not able to reproduce this issue.

    Make sure you are using MMWAVE_SDK:3.5.0.04 and other required components

    Thank you

    Cesar