Posts by gonca_santos11

Thank you for the response, the thread was really helpful. I just have a couple of questions.

I was trying your tip of using SPTP but I get the same results with or without $SPL_VEL_RESTR. Also the accelerations when using PTP are much slower compared to PTP.

Finally when you said to record the axis acc, if I am thinking correctly, this would only work if we kept the same conditions because if we changed the vaues of the axis or the velocities than the resulting acceleration would change. On the other hand this method can be used for the axis velocities because they don´t depend on the value of the axis. Is this correct?

Hello everyone,

I am working on Simulink with a KUKA KR22 R1610-2 and KRC 8.3.38. I am using PTP motions and I am having a hard time understanding how VEL_AXIS and ACC_AXIS work. For $VEL_AXIS when I use 100%, the constant speed is different from the one in $VEL_AXIS_MA. As for $ACC_AXIS from what I have been testing, it doesn´t seem to limit the maximum angular acceleration and I can´t find any logic in the way it affects the movement.

Thank you for your time,

Quote from panic mode

programming several linear motions on a same straight line and then turning them into one profile by simply adding C_DIS is not efficient - this can be done by single linear motion and without C_DIS. so not sure why or when one would need that.

I understand what you are saying but if you want to change the velocity and acceleration on specific points, I thought this was the only way

Thank you for your help,

Quote from panic mode

maybe motions are passing through or near singularity.

I don´t think that is the case because the acceleration only starts to decrease when I reduce the distance between motions.

I was just performing tests with collinear because I thought it was the simplest case, but what I want to do is perform a trajectory with linear and circular segments and to make the velocity change smoothly along that path, as a function of the position on the path. I am not sure what would be the best way to do that.

Hello all,

I am working with a KR22 R1610, KSS8.3.38 on Kuka.Sim Pro. I am trying to use SLIN motions with approximation to progressively increase the acceleration before reaching the final velocity, the code below serves as an example.

$SPL_VEL_MODE = #CART
SLIN {Y -1300}

$APO.CDIS = 10
$ACC.CP = 0.04
$VEL.CP = 0.1
SLIN {Y -1270} C_DIS

$ACC.CP = 0.06
SLIN {Y -1255} C_DIS

$ACC.CP = 0.08
SLIN {Y -1235} C_DIS

$ACC.CP = 0.1
SLIN {Y -1100}

The problem with this appproach is that, in some cases the acceleration drops before transitioning to the next acceleration and I don´t understand why this is happening.

Thank you all

DEF slinvel()
INI
$BWDSTART = FALSE
PFAT_ACT = PDEFAULT
FDAT_ACT = FHOME
$H_POS = XHOME
PTP XHOME

$SPL_VEL_MODE = #CART
$JERK.CP = 100
$ACC.CP = 0.2
$VEL.CP = 0.4
SLIN {X 100}

WAIT SEC 0.1
$TRACE.NAME[]="slinvel"
$TRACE.MODE = #T_START
SLIN {X 1300}
WAIT SEC 0.1
$TRACE.MODE = #STOP

END

The trace channel used was KRCIpo CartVEL_CmdIpo and the acceleration was obtained by calculating the difference quotient.

The test was performed with a KUKA KR22 R1610-2 using SIM PRO and Officelite with KSS V8.3.38.

Thank you for the help.

Quote from Fubini

What values are assigned to acceleration? What acc/jerk do you use for orientation?

In case of #opt axis specific values are used $acc_axis[], $gear_jerk[]. In case of #cart cartesian ones $acc and $jerk are used. This includes ori component.

Do you use inline forms or expert programs?

Also lookup system integrators manual. This stuff is explained in there and not the system variables documentation.

Fubini

Display More

I use the INI folder to start the program and the rest is all written by me.

I am using 100 for jerk.cp and 0.2 for acc.cp, I did not change the values of acc/jerk for orientation because i was testing for a linear path with no change in the orientation.

In the tests I have done the jerk is always bigger near the programmed velocity than when the speed is closed to zero which I dont understand.

Quote from Fubini

Similar trajectory planning but third order for slin can be activated by setting $spl_vel_mode to #CART.

I have tested both the $spl_vel_mode #cart and $spl_vel_restr in simulation (KUKA SIM PRO with officelite) and they seem to give the same results to each other so it left me wondering what their pratical difference is since in the system variables manual I could not get a clear explanation.

Also when using SLIN with #CART the time it takes to get to the programmed acceleration is much bigger compared to LIN, I have increased $JERK.CP which resulted in a slight improvement but I was wondering if there is some way to further reduce this time?

Thank you in advance,

Quote from SkyeFire

Velocity drop in corners is driven by inertia and servo maxima, not by LIN vs SLIN. SLIN is better at pre-plotting the path, but doesn't miraculously make the corner problem vanish.

I am working with KUKA officelite and SIM PRO, and in simulation when using C_DIS there is a clear difference between using LIN and SLIN, as SLIN is able to mantain constant velocity, the problem I have with SLIN is that the velocity profile is not so easy to control and is difficult to vary the velocity as a function of the path.

Hi everyone,

I have a KUKA KR22R1610_2 C4 with KSS 8.6.8

I am first defining a path, and then I need to make the velocity change smoothly along that path, as a function of the position on the path. The path may be defined with many points close together.

I have used LIN with $C_DIST to overlap velocity profiles between short distance points. However when points are not colinear, the velocity drops at the corner.

So now I am trying with splines. Or maybe there is a better way to do this?

Thank you all in advance!

I am doing my master thesis on mechanical engineering which involves using SLIN motions to perform a trajectory with any given velocity profile

Hello everyone,

I am working on a project that requires to discover the equations of motion for the SLIN motion, however i cannot find information about the velocity profile of this motion anywhere. Furthermore since it is only possible to define the maximum velocity in cartesian coordinates but not the acceleration, I cannot understand how the motion planner is able to create the trajectory with only the initial and final points and the maximum velocity.

Thank you in advance.