TM Robots with Standard Interface

Q&A
1

Hello. I am facing an issue in standard interface communication between Mechvision and TMFlow. I am using a Nano-Ultra Camera and a TM20X cobot.

I think it has something to do with Data Types but I will describe my issue in full below:

I have setup a Mech-Vision Project as per the backup here: backup_20250224134538_183.zip

I used a sample of general workpiece picking.

When I run the project via TMFlow in the actual robot, the vision project runs and seems to even show that parts are detected and an output string is generated (as seen in the logs below):

image
image1116×487 44.8 KB

However, the console shows an error of 1002: no vision result.

image
image740×222 10.5 KB

I believe I am just missing a critical step in formatting the output as per TM requirements.

I checked the error code page: Status Codes and Troubleshooting however for my error it is only giving a troubleshooting step as if the object was not detected, however, as you can see, the object is detected.

Please let me know if I have made any mistakes in my project.

PS: the backup I sent has output type “robot Path”, I have tried both “Robot path” and “Vision Results”

2

For the TMFlow Side, I’ve used the sample mm_vision_pick_and_place with the only modification being changing the IP address in the block: mm_v01_init1 and changing g_mm_point_number from 0 to 1 in the block: mm_v01_runvis1-> config (i tried with 0 and 2 as well)

3

In Mech-Vision project, please change your output format to vision result. If you use object matching, it should be like the image below.
When you use the sample mm_vision_pick_and_place, the robot just receives the visual picking points. you need to remove the path planning step from vision project.

image
image1414×455 27.2 KB

4

ahhh, so I dont need to use path planning from Mech-Viz? I will have to modify the pre-pick and depart positions myself in TMFlow? Thank you for the response