Slow robot movement in for loop | How to move fast through many close points

    Hello everyone!

    I am new to programming Fanuc Robots and I am struggeling with my rather specific setup. I have 10 PRs that can contain points. The number of points that I actually want to drive is not fixed, so it can be anything between 1 to 10 points. When I run this code the robot moves to all points but the more points I drive, the slower the robot movement gets (for all points!). The slowdown is really significant, so I was wondering if I am doing something wrong or if there is a better approach. What do you usually do if you have to drive through many points that are close to each other, to get high execution speed?

    Here is the code I have:

    R[36:adj nr of pts] = R[32:nr pts to drive] - 1


    FOR R[37:iterator]=0 TO R[36:adj nr of pts] ;

    R[30:curr pr] = R[37:iterator] + R[40:first pr] ;

    J PR[R[30:curr pr]] 100% CNT 100 ;

    ENDFOR ;


    Please let me know if I need to clarify anything, any help is greatly appreciated.


    Thanks!

  • Fubini March 19, 2025 at 11:04 AM

    Approved the thread.

  • decamargo March 19, 2025 at 11:29 AM

    Changed the title of the thread from “Robot movement in for loop slow” to “Slow robot movement in for loop | How to move fast through many close points”.

    Thank you for replying to the post! The points can be (and often are) very close to each other in cartesiean space (distance of a few mms to cms). As far as I understand, this heavily constrains the fanuc motion planner and therefore results in slower motion as the robot never reaches high speed.

    But I have two questions regarding this:

    1. How far does the fanuc motion planner "look ahead"? I would assume that the more points it can look ahead, the faster it can get. Is this how it really works and can I control how far it looks ahead?

    2. Is there a way to optimize the execution of continuous, short high speed motions? How do you usually approach this apart from "optimizing" the trajectory by getting rid of points? Can it make sense to reduce the overall speed or acceleration?

    Even if the Fanuc motion planner does look ahead, there is nothing to look ahead in your loop with an indirectly addressed target. The system simply does not know the future targets for ahead calculations, including the actual move CNT transition to the next move.

    Try listing these motions consecutively, with direct targets and no loop, and test for results.

    Do it well right away. It will become bad by itself.

    • New

    How many points do your paths contain? If the paths contain less than 100 or so points, you could set aside a block of sequential Position Registers and use a for loop to populate those PR's before executing a fixed block of code.

    Here is some TP Sudo-Code:


Advertising from our partners