AWR1843BOOST: AWR1843's degree accuracy

Hi Hiroyuki-san,

I am unable to understand how you have arrived at the 18 degrees as the angular resolution (What are 1.79 and 3.58 degrees?).  Are you saying you are seeing a shift in the angle measured? It would be very helpful for me to understand this if you could provide me a pictorial depiction of your problem.

Thanks & Regards,

Kaushik

As a procedure, the two objects used this time are aluminum plates of 30 x 6 cm^2.

When I measured it once with an aluminum plate, it was detected with a deviation of 1.79 to 3.58 deg. (perpendicular to the millimeter-wave radar antenna plane)

(Figure 1)

Next, when I measured two aluminum plates with an interval of 14.5 deg, the result was 18.1 deg. When I saw it, I thought that the angle accuracy was out.

(Figure 2)

The figures is shown below.

Figure 1

Figure 2

Hi Hiroyuki-san,

Can you clarify the following points:

• Have you calibrated the sensor? If not, kindly request you to do the same.
• In your Figure (2), is the angle between the two objects with respect to the sensor 14.5 degrees or is it the DOA with respect to the Centre of the radar's field of view? If it's the latter, are you seeing this offset in both measures?
• May I know how you are measuring the angles using the sensor? (OOB + Visualizer, custom algorithm etc.)

Regards,

Kaushik

Thank you for your replying!

・Have you calibrated the sensor? If not, kindly request you to do the same.

What kind of work do you specifically refer to as proofreading?
(Threshold, etc.?)

In your Figure (2), is the angle between the two objects with respect to the sensor 14.5 degrees or is it the DOA with respect to the Centre of the radar's field of view? If it's the latter, are you seeing this offset in both measures?

I personally think of it as the angle between two objects relative to the sensor.
(Shown the figure.)

・May I know how you are measuring the angles using the sensor? (OOB + Visualizer, custom algorithm etc.)

Measured with Visualizer with 3 transmitting antennas and 4 receiving antennas. (file attached)

By the way, the data recorded there was converted to CSV, and the angle was calculated from those with the same range size in the same frame.

Sample.CFG

Hi Hiroyuki-san,

Thank you for clarifying.

Hiroyuki Kanematu said:

What kind of work do you specifically refer to as proofreading?
(Threshold, etc.?)

I'm unable to understand your question here. Could you please rephrase the same? Once again, I request you to perform calibration on the sensor.

Also, can you send the point cloud CSV that you are using.

Regards,

Kaushik 

OK, I'll send to you.

(Two type csv file)

①.csv is figure1 type, ②.csv is figure2 type②.csv.

①.csv

Hi, Thank you for your replying.

This file is right.

I attached this file.(from 56~66 frame)

3058.②.csv

And I watch your topics thank you!