# KRC4 spline vs. KRC2 Linear? Quick question

• Hey everyone,

I'm looking to upgrade to the KRC4 from the KRC2 system and I was curious if anyone knew if there are any speed/velocity differences between the way the software interprits commands. I know most of the speed is dictated by the robots themselves, but I am curious if a P2P/Linear command would be interprited smoother or at a higher rate of speed with the KRC4 P2P/Linear or possible Spline commands. P2P commands move very quickly on our current system, where Linear commands seems to be significantly slower in regards to the robots maximum velocity of 3m/s.

Cheers, and thanks ahead of time for the information

• On KRC4 you have two motion planners and you don't want to mix them (every time you do, robot pauses so if you have to switch, do it at point where robot stops anyway).
Think:
MP1 = old (legacy) motion planner, with instructions PTP, LIN, CIRC
MP2 = new (spline) motion planner, with all other motion instructions (Spline, CPTP, SLIN, SCIRC, S-whatever...).

CP motions in MP2 are significantly faster than CP motions in MP1
PTP motion in MP2 is (very) slightly slower than in MP1

2) if you have an issue with robot, post question in the correct forum section... do NOT contact me directly

3) read 1 and 2

• Any interpolated motion (LIN, CIRC, Spline, etc) is going to be slower than any PTP motion between the same two endpoints. That's because PTP motions are always speed-optimized, but interpolated motions are not -- they're constrained by other factors, like keeping the TCP moving along a pre-calculated path.

Imagine a simple motion from {X 1000, Y 1000, Z 1000} to {X 1000, Y -1000, Z 1000} carried out in \$ROBROOT -- it's a simple motion from left-to-right in front of the robot. For a PTP motion, the robot simply calculates how much to rotate each axis, figures out which axis needs the most time to move the distance, and slows the other axes down to match that time.
For a LIN motion between those points, however, A2 and A3 will have to reverse direction as the TCP passes through Y=0. Since each axis needs time to slow down, stop, and accelerate in the opposite direction, the motion planner has to take those accel/decel limits into account while generating the motion.

Then, of course, there's singularities, which just make it impossible for some axes to keep up.

For LIN vs Spline... I haven't used Spline on KRC4s yet, but I've heard that it may now be superior to "classic" LIN motions in speed, accuracy, and smoothness. It does behave differently from "classic" LIN motions, though, so you need to be careful about blithely swapping one command for the other.

## Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account
Sign up for a new account in our community. It's easy!