Question related to hardware

Hello,
In the following course most of the PID and inverse dynamic control is under assume that the control input is the voltage of each joint. on many real robotic application and motor the control input is position and velocity. how the method talked about in this course can be implement when the input is position/velocity/accelration?

what will be the v vector for the inversedynamic?
what will be also the PD control?

thank you for any answer :slight_smile:

Hello @noam92,

you are right, when the control input is in terms of position, velocity, or acceleration, the controller design must be modified. You could for instance use a feedforward controller to generate the desired trajectory in terms of position, velocity, and acceleration, and then use a feedback controller to track that trajectory.

For the inverse dynamics controller, if the control input is the joint position and velocity, the v vector has
to be modified to include the joint acceleration. Specifically, the v vector has to contain the joint acceleration term along with the joint velocity and position terms.

For the PD controller, the control law has to be modified to take into account the joint acceleration. The P and D gains are multiplied by the position error and the velocity error, respectively, while the acceleration error has to be added to the control output.

At the moment we have not covered this topic in the academy, not because it is unimportant, but because it is a big topic on its own.

Hope this helps,

Roberto

1 Like

This topic was automatically closed 5 days after the last reply. New replies are no longer allowed.