3
$\begingroup$

I am studying The Modern Robotics: Mechanisms, Planning, and Control (by Kevin Lynch, Frank Park and Evan Suma). I am trying to implement Chapter 11.7.1 Impedance-Control Algorithm of the book with ROS simulation, but I am confused and I am asking for help.

he impedance control algorithm in the book has a dynamics formula for task space, but I am confused as to whether the theta used here is the desired theta or the current theta. In addition, jacobian and twist are also used to create Lambda and Eta, and I am not sure whether they are also desired jacobian, desired twist, current jacobian, or desired twist. Could you please help?enter image description here

$\endgroup$

1 Answer 1

3
$\begingroup$

All of the $\theta$'s listed represent current joint angles, because we are considering what the robot's dynamics look like at this moment in time - i.e. at the given configuration. Since this is an end-effector force control law, I would imagine the only "desired" values could be those of $x$, $\dot{x}$, and $\ddot{x}$ in the $f_{\text{ext}}$ term.

As for $\tilde{\Lambda}$ and $\tilde{\eta}$ - because these are the approximated equations of motion for your robot arm in the task space, they use the actual twist $\mathcal{V}$ and actual Jacobian $\mathbf{J}$. As a general rule of thumb, if you are using a model to "compensate" for your system dynamics, you will be using the true system state. Otherwise, you would be tracking, minimizing, etc. some sort of error associated with a desired value.

$\endgroup$
2
  • $\begingroup$ Thank you for your help! Thank you for clearing up the confusion I had. $\endgroup$
    – HARCO
    Commented Feb 8, 2023 at 4:49
  • $\begingroup$ No problem! Best of luck, it looks like you're working on interesting stuff! $\endgroup$ Commented Feb 8, 2023 at 17:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.