Scribd
Upload
57 views

Computed Torque Control: by Vasfi Emre Ömürlü, PH.D

This document discusses computed torque control for robotics. It describes using a dynamic model to represent a robot as a set of differential equations and designing a computed torque controller to cancel out nonlinear effects like gravity, friction, and inertia. A PD gravity controller is also designed to compensate for gravity loading. Examples are provided of implementing computed torque control and PD gravity control for single and multi-axis robots.

Uploaded by

_MerKeZ_
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
57 views

Computed Torque Control: by Vasfi Emre Ömürlü, PH.D

This document discusses computed torque control for robotics. It describes using a dynamic model to represent a robot as a set of differential equations and designing a computed torque controller to cancel out nonlinear effects like gravity, friction, and inertia. A PD gravity controller is also designed to compensate for gravity loading. Examples are provided of implementing computed torque control and PD gravity control for single and multi-axis robots.

Uploaded by

_MerKeZ_
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Computed Torque Control

By Vasfi Emre mrl, Ph.D.


- Use Lagrance Euler Dynamic model to
control a robot
- Represent a dynamic model as a set of
first order diff. eq.
- Design a computed torque controller
- Design PD gravity controller

By Vasfi Emre mrl, Ph.D., 2005 2


Manipulator State Eq. Model
For an n link manipulator, this is a set of n
second-order nonlinear differential equations.

Note: D-1(q) exists

Let xT =[qT,vT] where v = q then above


equation can be written as a set of first order
diff. eqs.

By Vasfi Emre mrl, Ph.D., 2005 3


Computed-Torque Control (Model
Based Control)
Feedback asignal that cancels the effects of gravity, friction,
the manipulator inertia tensor and Coriolis and centrifugal
forces. This cancellation will be done by using the
Lagrange-Euler Dynamic Model.

All of these effects were treated as disturbances in the


single-joint controller.

To cancel the nonlinear effects, consider the feedback law,

By Vasfi Emre mrl, Ph.D., 2005 4


By Vasfi Emre mrl, Ph.D., 2005 5
If the estimates are exact, what are
the closed loop dynamics?

By Vasfi Emre mrl, Ph.D., 2005 6


Example: implement the computed-
torque law for the two axis planar robot

By Vasfi Emre mrl, Ph.D., 2005 7


Computed-Torque Control
Accounts for the dynamic interactions among the
links unlike using a control loop around the joint.
Try to cancel the nonlinear terms due to gravity,
inerti, Coriolis and centrifugal forces

By Vasfi Emre mrl, Ph.D., 2005 8


Example

Find a computed-torque control law for the


one-axis robot with a load at the tip, mL

By Vasfi Emre mrl, Ph.D., 2005 9


PD gravity controller

By Vasfi Emre mrl, Ph.D., 2005 10


Examples
Implement a PD gravity controller for the two axis planar
robot

Using the equations of motion fromt the computed torque


control law example, identify the gravity loading terms.

By Vasfi Emre mrl, Ph.D., 2005 11


Examples
A three axis SCARA
robot,
Dynamic model for
the third axis is

Write the state eq.


For the dynamic
model of joint 3.

By Vasfi Emre mrl, Ph.D., 2005 12


PD gravity control law for joint 3 is
given by

With this controller, find the characteristic eq. of the closed loop
system

Find the gains in the controller so that the closed loop system will
have a damping ratio of sqrt(2) and =1 rad/sec

By Vasfi Emre mrl, Ph.D., 2005 13

You might also like