IWR1642BOOST: People Counting Demo Crashes: Seeking Troubleshooting Methods and Diagnostic Questions

Hi Raymond,

It looks like there is an error in the default configuration. The ES2 devices are required to run in low power mode, so the line lowPower 0 0 should be lowPowe 0 1.  You can find other configurations in <Industrial_Toolbox_2.5.1>/chirps/images/People+Counting+and+Tracking  with the correct lowPower call.

If this does not fix the issue we can discuss a debug plan.

Regards,

Justin

The .bin file was reflashed: C:\ti\mmwave_industrial_toolbox_2_5_1__win\mmwave_industrial_toolbox_2_5_1\labs\lab0011-pplcount\lab0011_pplcount_quickstart\xwr16xx_pcount_lab.bin

A config file was selected from Visualizer UI:

C:\ti\mmwave_industrial_toolbox_2_5_1__win\mmwave_industrial_toolbox_2_5_1\chirps\images\People+Tracking+and+Counting+Application\mmw_PC_6m_ConferenceRoom.cfg

Feedback indicates the .cfg file was applied.:

Capture before ~8 seconds before crash:

Not sure why the configured power setting does not show: Capture just before crash:

Questions:

The error message indicates the application is crashing because of a timeout:: not receiving expected amount of data.  It would not surprise me that the index that exceeds the matrix dimensions, is the root cause of the problem.  

1)  Does it make sense to troubleshoot based on the warning messages?

2)  Did the Subject Matter Expert reproduce the warning messages? 

3)  Is the root cause of the isssue something I have \ have not done?  Maybe there is a checklist of diagnostic questions to check my configuration \ process?

Thanks for explanation: the utility of responses is improving.

A config file was selected from Visualizer UI:

C:\ti\mmwave_industrial_toolbox_2_5_1__win\mmwave_industrial_toolbox_2_5_1\chirps\images\People+Tracking+and+Counting+Application\mmw_PC_6m_ConferenceRoom.cfg

dfeDataOutputMode 1
channelCfg 15 3 0
adcCfg 2 1
adcbufCfg 0 1 1 1
profileCfg 0 77 30 7 62 0 0 60 1 128 2500 0 0 30
chirpCfg 0 0 0 0 0 0 0 1
chirpCfg 1 1 0 0 0 0 0 2
frameCfg 0 1 128 0 50 1 0
lowPower 0 0
guiMonitor 1 1 0 0
cfarCfg 6 4 4 0 0 16 16 4 4 50 62 0
doaCfg 600 1875 30 1
SceneryParam -6 6 0.05 4
GatingParam 4 3 2 0
StateParam 10 5 10 100 5
AllocationParam 250 0.01 5 1 2
VariationParam 0.289 0.289 1.0
trackingCfg 1 2 250 20 200 50 90
sensorStart

QUESTIONS

Is said .cfg file intended for the people counting exercise? If not, please advise the path\filename.  

Did the SME provide an explanation as to why the power setting cause a failure to return data?

Thanks,
Raymond

Hi,

I am also facing the same problem.

Please let me know if you get anything concrete on the issue. A few days back it was working perfectly but since last 2 days, it is creating a problem. I have not changed anything on my computer or CCS. 

Interesting Laxmi. Although we are both using  mmwave industrial toolbox 2.5.1, your visulaizer app worked at one time.

@Laxmi, can you post your .cfg file at the bottom of your original post?  I'd like to compare yours with my default.cfg file.  I have posted my setting above in a previous post.

Hi Raymond and Laxmi,

In both of your screenshots, the Power mode is incorrect. Low power mode needs to be enabled, or the device will not run.  Please change the configuration to lowPower 0 1.

ES2.0 devices only work in low power mode, however, the option is still present in the SDK to run in either low or high power mode. 

Regards,

Justin

.cfg Power setting was update and uploaded to EVM.  Visualizer crashes.  Error messages and confirmation of 'lowpower' settings:

What other steps \ actions should be tried to successfully run the people counter demo?

Attempts to perform the above suggestions have failed.   Yes,  The Uniflash instructions indicate that after a successful flash to Power off the board and remove the SOP2 jumper.  I reperformed the entire exercise (starting with flashing the TI compiled binary) and reset the EVM with SW2 (in yellow box), just before sending the .cfg file from the visualizer.   

Only TI pre-compiled binaries are used.  I have not compiled any binaries for the EVM.

Serial monitoring programs (Termite \ Putty) do not show activity on the COM5. After flashing, removing power + SOP2 jumper, and rebooting, should I expect the EVM to return data that woudl be visible on Putty \ Termite?  Or does the configuration file and start need to be invoked first?   What should the serial port settings be?  I tried representatives setting of 9600 and 115,2K

Justin,

We have tried these diagnosis / corrections to address the people-counting visualizer crash:

1. Updating the  configuration file with updated low power settings
2. "listening" for data with terminal programs (Putty \ termite)

The tests indicate the TLV data is not being sent by the device.  Is there any other tests to perform or is it time to RMA the EVM?

Thanks,

Raymond

Hi Raymond,

From the error shown in your previous post on running the Out of Box Demo, the analog Monitor configuration is failing. When running the OOB on SDK 2.0.0.4, did it show the same error as the one you posted here? 

These errors give line numbers, which allow the user to debug the demo by checking the source code at that line to determine why the error was thrown. 

Regards,

Justin

Hi Justin,

The OOB demo SDK 2.0.0.4 observations are posted at https://e2e.ti.com/support/sensor/mmwave_sensors/f/1023/t/717129

The EVM is no longer in my custody and I am unable to perform any more tests.   That being said, it is interesting that the line numbers provide insight as to the source code line to be inspected for errors: I did not compile any code and flashed only the TI provided .bin files (KISS principle).

Your diligence is appreciated.

Thanks,

Raymond