IWR6843: Locate static objects

Hi Daniel,

I've asked an mmWave Expert to respond to your query and someone should have an answer for you in the next few days.

Cheers,

Akash

Hi, Daniel:

I am Zigang specialized on radar systems.    Please let me know what exactly you are looking for to improve. 

Best,

Zigang 

Hello Zigang.

We design IoT devices for smart cities. Some of our devices are designed to measure the available capacity of garbage bins, now we use ultrasonics as a distance sensors.

Right now we are using a IWR6843ISK-ODS with the MMWAVEICBOOST. Our objetive is to measure the profile of a garbage bin, in order to know how full is. All the objects will be static, so we need a Chirp Profile in order to measure the distance and something like a mapping of the surface of the garbage. Also will be appreciated some info regarding the best Industrial toolbox app to achieve the goal.

The first test have been made in a little garbage bin of about 40x40cm and the garbage level at about 70cm away from the sensor, the garbage bin is made out of ABS and some pieces of cardboard, plastic and metal parts as garbage.

Is it possible to to detect those pieces? Is it possible to detect the inner walls of the garbage bin?

You can find pictures of the setup on the original topic here:
e2e.ti.com/.../809626

Thanks for your time.
Best Regards.

Hi, Daniel:

It is very challenge to detect the inner wall of garbage bin.    For garbage bin, for a giving range, the sensor receives many different reflection from different angle.   These multiple reflection can not separate from range profile, and has to rely on angle of arrival estimation.   But single chip radar like ODS antenna does not have very high angle resolution to separate objects that are close in angle and belong to the same range bin.   

Right now, we are developing a static object detection lab.  The main idea is to record the static background scene (in your case, it is the empty garbage bin).  Then we will calculate the delta of the current scene with the recorded static background, and then detect the added on objects.   This approach may have better chance to detect the internal garbage.   One drawback of this method is the background scene can not change too much.   If the sensor location kinds of moving over time, then the detection sensitivity may reduce or there can be some false detection.   Let me know whether this lab sounds interesting to you.   This lab is in testing, will be available soon.   

For the chirp design, I would recommend to use the full 4G bandwidth to minimize the range resolution.   And use smaller number of ADC samples and larger number of chirps per frame.    

Best,

Zigang

Another lab that may be interesting to you is the high accuracy lab, which trying to measure the liquid level in a gas tank.   This lab uses a horn antenna, and make sure the closest reflection comes from the liquid reflection.  And the body of the tank will not be in the antenna field of view. 

Best,

Zigang

HI, 

I will close this thread if I did not receive any feedback from you.

Best.

Zigang

Hello,

Can you please wait some days more? I had no time to test your advices, just to have the chance to ask again without the need to open a new thread.

Do you know how long can be the static object detection lab?

Best regards.