Low level Control

[jnc]

Hi..

I am going to work with our new Tx60 & CS8C controller to be delivered at the end of this year, hopefully.

I wonder if the robot system still has the feature called 'Low Level Control' (LLC) or 'Low Level Interface' (LLI). According to my search on web, some people utilised the feature to feed forward the current (or torque) to the arm. Since I am interested in the motion control part, the feature is important to me.

So, if any one knows about the feature and it is still supported by the Staubli, can you just give a short reply?

Thanks in advance.
jnc.

[Jim C]

Low Level Interface or LLI is still available. This option should be ordered with system. It is an alternate operating system for the controller that removes the VAL 3 programming language. Instead you program in C. The key here is low level. The LLI API gives you access to the servos, IO (including e-stop), and to a forward and reverse kinematic solver. In your C applications you have to write your own code for motion control and trajectory planing.

The indented customer for LLI is universities, colleges, and R&D. It for people who want to write trajectory planing and motion control software.

The initial purchase price is high (I am not in sales and don't quote prices talk to your sales guy). It include telephone support from R&D to get you started. The purchase price for additional systems  is much lower, but does not include R&D support.

Staubli can provide maintenance support for LLI arms. We boot the system up from USB using a maintenance version of our standard software. You are pretty much on your own for application support since you are not using VAL 3.

LLI gives you a few methods of servo control including a torque control or a position control.

Standard VAL 3 does give you an instruction, getJointForce which allows you to read the current joint torque (calculated based on motor current, gear ratios, and joint type). Often this instruction will be used to check things like insertion force during assembly. If the values from getJointForce is higher then expected, the movement can be stop and the arm can then be moved back from the part. However the servo control remains positional. 

[jnc]

Thank you very much for your wonderful remark, Jim.

I may have to think seriously if I really want LLI..

Thank you again.
jnc.

[Jim C]

What exactly are you trying to do with the arm?

[jnc]

What I am trying to do is to develop robot force control algorithms. (Currently there is no specific project on duty using the robot.)  I am working at a university and need to build something new and innovative. These algorithms very often require the knowledge of robot dynamics and force measurement. Of course, the availabilty of torque servo control in the Staubli system is most desirable to me. 

However, based on your remark, I am not sure if I can succeed the LLI based development by myself. (Yah.. I am a little scared. ) I am acquainted with motion control, some spimple hardware setup, robot kinematics/dynamics, a little on Real time OS, but don't know if the level of my knowledge is sufficient to go for LLI. And  the time needed... Honestly I am a little lost right now.. 

Thanks, Jim, for asking.
jnc

[Jim C]

JNC send me an e-mail, j.cook@staubli.com, and I can send you some more information. I cannot just post it. R&D controls this product and wants information on it to be limited. LLI is not for everyone, but it might be a good solution for you.