• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 333 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.

Original question:

IWR6843AOPEVM: Long Range People Detection

IWR6843AOPEVM: Improve the Tracking Lab on 6843AOP EVM

radar_guru
Prodigy 70 points
Part Number: IWR6843AOPEVM
Other Parts Discussed in Thread: IWR6843AOP, IWR6843ISK

Tool/software:

Hello Team,

In the People Tracking Lab for the IWR6843AOP, I can track pedestrians effectively up to a range of 20 meters. However, I aim to extend this range to 50 meters. I understand that the beamforming feature is not available in AOP EVMs.

Upon comparing the OOB (Out-of-Box) Lab and the Long Range People Detection (LRPD) Lab, I noticed that the configuration commands are the same, with the LRPD Lab incorporating a tracking layer implementation. I achieved a 50-meter range in the OOB Lab; however, when using the same detection layer parameters and CLI commands in the LRPD Lab, I encounter an "out_of_index" error.

Additionally, the People Tracking Lab performs well for detecting pedestrians up to 20 meters, but the tracking is unreliable when applied to cyclists or cars. Could you advise on how I might improve the tracking of cars using this configuration?

Many thanks,

Jomin Joy

  • 0
    TI__Mastermind 37805 points
    radar_guru said:
    Upon comparing the OOB (Out-of-Box) Lab and the Long Range People Detection (LRPD) Lab, I noticed that the configuration commands are the same, with the LRPD Lab incorporating a tracking layer implementation. I achieved a 50-meter range in the OOB Lab; however, when using the same detection layer parameters and CLI commands in the LRPD Lab, I encounter an "out_of_index" error.

    Please use a config file designed for the LRPD demo when running the LRPD demo. These can be found in the LRPD folder of the radar toolbox.

    radar_guru said:

    Additionally, the People Tracking Lab performs well for detecting pedestrians up to 20 meters, but the tracking is unreliable when applied to cyclists or cars. Could you advise on how I might improve the tracking of cars using this configuration?

    See related questions. One good thread is here.

    https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/804110/iwr6843isk-iwr6843isk---cars-detection

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 70 points

    Hi Nathan,

    Thanks for your reply.

    I am planning to implement a 50-meter range for the people tracking lab. However, when I modify any parameters in the profileCfg CLI command, the lab stops working. Could you please advise on achieving this range with the 6843AOP?

    The LRPD is available only for ISK modules provided in the link: https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1331490/iwr6843aopevm-long-range-people-detection. 

  • 0 in reply to Nathan Block
    Prodigy 70 points

    Hi Nathan,

    Most of the labs including LRPD, Traffic Monitoring, Mobile Tracker are implemented for 6843 and 6843ISK modules. Could you please provide a link on how to change the antenna geometry values in the binaries to work with 6843AOP?

  • 0 in reply to radar_guru
    TI__Mastermind 37805 points

    Hi,

    My apologies - the LRPD lab isn't designed for the AOP device because the AOP's antenna geometry does not support TX beam steering.

    If you start with a working configuration file for the people tracking lab and ONLY modify the frequency slope, you should get increased detection range. What parameters are you modifying in the profileCfg?

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 70 points

    Hi Nathan,

    I'm modifying the following CLI parameters:

    profileCfg 0 60 35 6 43 0 0 14 1 125 3433 0 0 48
    chirpCfg 0 0 0 0 0 0 0 1
    chirpCfg 1 1 0 0 0 0 0 4
    chirpCfg 2 2 0 0 0 0 0 2
    frameCfg 0 2 96 0 100 1 0
    As per the mmwave estimator the range of detection should be 30m but I could get a detection range of 18m only. Could you please help me to resolve the issue. Many thanks.
  • 0 in reply to radar_guru
    TI__Mastermind 37805 points

    Hi,

    Do you know if you have enough link budget to detect at 30 meters away? See the radar equation (2) here. If the problem comes from link budget, then you may sometimes detect at ranges greater than 18 meters, especially for large targets, like moving cars. Can you confirm that this is the case?

    If the problem comes from link budget, then you can increase your link budget by increasing your chirp time or number of chirps. To increase the link budget by 30/18 = 1.67 -> 67%, you will need to increase your total chirping time by a factor of 1.67^4 = 7.7.

    In general, I think you need to change your device to an IWR6843ISK and just use the labs that are designed to give you 30 meters of detection range. You are trying to accomplish this with software that was not designed to do so, and it would be much easier for you to simply use the software and hardware designed to accomplish what you're trying to do.

    Best,

    Nate

  • 0 in reply to Nathan Block
    Prodigy 70 points

    Thanks Nathan.