Scribd
Upload
129 views11 pages

Examples of Automatic Control Systems

Automatic control systems can be open-loop (feedforward) or closed-loop (feedback). Early examples include the Jacquard loom (1801) which used punched cards to program patterns without feedback, and Roman aqueducts which used floating valves controlled by water levels. Feedback systems became more common and important. Control systems are useful for processes that are impossible, difficult, costly or unsafe for humans to control directly due to factors like hazardous environments, fast responses or constant operation. Theoretical analysis of control systems was pioneered in the 19th century and enables computerized industrial and animation applications today. Closed-loop systems using feedback are now preferred over open-loop systems. State machines can model cyclic human behaviors like walking

Uploaded by

Đăng Trọng
Copyright
© Attribution Non-Commercial (BY-NC)
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
129 views11 pages

Examples of Automatic Control Systems

Automatic control systems can be open-loop (feedforward) or closed-loop (feedback). Early examples include the Jacquard loom (1801) which used punched cards to program patterns without feedback, and Roman aqueducts which used floating valves controlled by water levels. Feedback systems became more common and important. Control systems are useful for processes that are impossible, difficult, costly or unsafe for humans to control directly due to factors like hazardous environments, fast responses or constant operation. Theoretical analysis of control systems was pioneered in the 19th century and enables computerized industrial and animation applications today. Closed-loop systems using feedback are now preferred over open-loop systems. State machines can model cyclic human behaviors like walking

Uploaded by

Đăng Trọng
Copyright
© Attribution Non-Commercial (BY-NC)
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

Automatic control systems 2/15/05 10:07 AM

Automatic control systems

Feedforward - open loop systems,


• early example: Jacquard loom of 1801

•
• set of punched cards programmed the patterns to be woven by the
loom, and no information from the process or results was used to
correct the loom operation.

Feedback – closed loop systems


• systems feed back information from the process to control the
operation of the machine
• earliest closed loop systems was that used by the Romans to
maintain water levels in their aqueducts by means of floating valves

•
• windmills were the spawning ground of several control systems,
for example the sails were automatically kept into the wind by
means of a fantail (1745), as shown below; centrifugal governors
were used to control the speed of the millstones (1783), and the
speed of rotation of the sails was automatically controlled by roller
reefing (1789)
•

Why control systems?


• many variables can be controlled by humans, however:
• in practice this may be
o impossible/difficult
o costly
o undesirable because of the need for continuous operation in a
hazardous environment (large forces, fast responses, etc.)
• human reaction time is about 0.3 seconds – too slow!
Theoretical analysis of control systems
• first published by Maxwell, 19th century

•
• theory enables computerized control
o industrial applications
o animation applications

Feedback is essential!
• Closed loop systems are the way to go
• Block diagram of a generic closed loop system:

•
• The output can affect the input because of the feedback loop.
• The Appendix to these notes contains an example of the analysis of
a simple control system.

• Higher level control and state machines


o Consider a running human being
o Model the human as several rigid pieces connected by hinges
 this is an articulated figure, and we will study this more
as the class progresses

•
• Prof. Jessica Hodgins (now at CMU) led many of the developments
in designing controllers to simulate human activities using
dynamically-driven articulated figures.
• People are very attuned to the subtleties of human appearance and
motion, so it’s a challenge!
• Let’s divide this task into a hierarchy of controls
o LOW LEVEL – control each joint servo (a servo is a small
motor that applies torque to a joint)
o MID LEVEL – control each phase of the gait (a gait is a
person’s manner of walking)
o HIGH LEVEL – determine where the person should run
• How do we achieve these?
o LOW LEVEL – simple closed-loop controller that has a desired
angular position or angular velocity and measures the current
angular position/velocity.
o MID LEVEL – state machines, as we discuss below
o HIGH LEVEL – Demetri Terzopoulos will discuss this on
Thursday
• First, let’s talk about what we are trying to control, i.e., the human
body
o a bunch of “rigid” pieces
o connected by joints

o
o sometimes knees and other joints can store energy (in
tendons), acting as a hinge spring
o we need to know some data about the body

o
State machines
• running is a cyclic behavior
• at each “stage” the muscles have different roles/responsibilities
• some parts of the body may be active (i.e., a stance leg) or passive
(i.e., a swinging leg).
o The active parts achieve the desired motion,
o but the passive parts play a key role too: they move so as to
reduce the overall disturbance on the body.
• State machines may be represented in a block diagram:
•
• or they may be represented in a table:
•
2/15/05 10:07 AM
2/15/05 10:07 AM

You might also like