With the robot Commando with Engineer Kit 1 + Kit 2, we are able to use the default line detection function with Raspi Zero W + Camera by setting the camera mode to 5 and set the sub mode to the default color (such as Red color, etc). However, how to detect black color?
We found that there are H, S, V setting, however when we set the value, the line detection is still detecting the color in sub mode. How to let the detection is detecting the colour with these H, S, V setting?
Very good question! So far, ROBOTIS has not released the detailed control table for the ROBOTIS’s RPi0W+Pi Camera, so I am not sure myself how the min-max values of HSV relate to the SubMode (Addr. 55), when the Mode (Addr. 11) is set Color Detection (Mode=1). Furthermore, I do not have the ROBOTIS Pi-Camera handy to test right now, so below are just my “best guesses” and they are not guaranteed to work correctly, so you can try them and report back on this post and we’ll see.
First, what kind of min-max values for the HSV variables did you use? Did MANAGER tool let you save them in memory?
In the HSV Color Space (which is 3-dimensional), the Black-to-White objects are located along on the Central Vertical Axis (see this Wikipedia article HSL and HSV - Wikipedia). This means that you should try to set the various min-max HSV values as follows:
Then you still need to set the Sub-Mode (Addr. 55) to some kind of value! And here is my “wild guess”, choose “White”, i.e. “7” for “Sub-Mode”. If this scheme works, I’ll put this solution in the second edition of my “Engineer” books.
Another “lateral thinking” approach is also possible: why don’t you have a black background and use “white” tape to make the track for the robot to follow?
Thank you very much for your reply.
MANAGER tool does not let me edit the HSV value, but we can load the value with codes in R+Task3.0. After that I checked in the R+Manager, all the value is written. However the camera is still detecting the colour set in the Sub Mode!
I run a task code to test on changing the Sub Mode from (255 ~ 0), like this:
But the robot only play melody when the Sub Mode is set to 7 (White color) but it is not detecting the black, it is detecting the white colour…
We cannot change to white line in black background because black line and white background is the setup of the competition.
Any idea which data should be set to let the Raspi use the HSV value to detect line?
It looks like if you stay with this setup: using the RPi0+Pi Camera as a DXL 201 and as designed by ROBOTIS, it does not give you enough flexibility to track a Black Line.
From now on, you have several options depending on your current hardware/software knowledge:
Then you can use the RPi camera to detect other objects that it is designed to detect using ROBOTIS firmware.
https://www.amazon.com/Projects-Guide-ROBOTIS-ENGINEER-Combined-ebook/dp/B09KQCH2FV
https://www.amazon.com/Using-ARDUINO-ROBOTIS-Systems-Ngoc-ebook/dp/B0BPXGQ6YX
You have all these books already right? From Summer 2024 meeting?
Thank you for your reply. 
I try to stay with this setup: “using the RPi0+Pi Camera as a DXL 201 and as designed by ROBOTIS", because it is easier for the students to code in Task and handle the hardware.
I got your book, so I will study and consider your suggested setup. Thanks again. ^^
Use the Dream kits frame parts to mount the IR sensors and 3M Dual Lock tapes to position them any where needed.
dear experts,
can we use two mode in a code? for example i want to detect color and at the same time i want to stream the video.
The camera can maintain only one mode at any one time. But in theory you can switch mode within the same program. And it takes a long time to switch mode back and forth.
If you use a car bot I would stop the car first before doing video streaming for a short time then switch to color tracking mode before letting the carbot move again also for a short time. Then your program can repeat this sequence at the next iteration So the overall performance will not be very good, unfortunately!
tq sir for your explaination. it seem that not much we can do using the default component n configuration (RPI0 + CM-550 + R+ Task).
i would like to
but it seem not able by using the default setting n component.
To do all that you will need a more powerful setup like with an OpenRB-150 combined with an RPi5 with its own RPi Cam and/or Husky Lens Gravity camera and then you also will need to program in C++ and Python too. And probably two separate cameras too.
Or you can set up two CM-550s with two Pi Zero on the same robot too!! Each CM-550 can do its own job while communicating to each other via RC-100 packets and BLE, or even wired serial comm through the 4 pins UART ports.
