Add two MRAC siso examples by KybernetikJo · Pull Request #914 · python-control/python-control

KybernetikJo · 2023-06-15T23:01:52Z

direct mrac using mit rule
direct mrac using lyapunov rule

examples based on Aström & Wittenmark

murrayrm · 2023-06-19T21:04:48Z

This PR will break once #916 is merged. The changes needed to use the new I/O system structure ar straightforward (eg, replace LinearIOSystem with StateSpace or better yet ss). I suggest we hold off on merging this until after #916, but running unit tests now to make sure there are no problems under various configurations.

coveralls · 2023-06-19T21:57:08Z

coverage: 94.849%. remained the same when pulling f3713f1 on KybernetikJo:mrac-examples into d153035 on python-control:main.

murrayrm

This PR needs to be rebased on top of the current main branch to catch the changes in #916, which eliminates the LinearIOSystem class and adds a ct.nlsys factory function. In addition, it would be good to replace calls to the class constructors (NonlinearIOSystem, InterconnectedSystem, etc) with calls to factor functions (ct.nlsys, ct.interconnect, etc).

murrayrm · 2023-07-02T16:14:27Z

doc/examples.rst

+   mrac_siso_direct_mit_rule_statespace
+   mrac_siso_direct_lya_rule_statespace


Perhaps use shorter filenames? It seems like mrac_siso_mit and mrac_siso_lyapunov would fit the bill.

murrayrm · 2023-07-02T16:15:10Z

doc/mrac_siso_direct_lya_rule_statespace.rst

@@ -0,0 +1,15 @@
+Model-Reference Adaptive Control (MRAC) siso, direct lyapunov rule


I would change "siso" to "SISO" (since it is an acronym) and capitalize Lyapunov (since it is a proper name).

murrayrm · 2023-07-02T16:15:37Z

examples/mrac_siso_direct_lya_rule_statespace.py

@@ -0,0 +1,185 @@
+# mrac_siso_direct_lya_rule_statespace.py
+# June 2023


Add author information?

murrayrm · 2023-07-02T16:17:14Z

examples/mrac_siso_direct_lya_rule_statespace.py

+G_plant_ss = ct.StateSpace(A,B,C,D)
+
+io_plant = ct.LinearIOSystem(
+    G_plant_ss,
+    inputs=('u'),
+    outputs=('x'),
+    states=('x'),
+    name='plant'
+)


This can all be done by a single call to ct.ss:

io_plant = ct.ss(A, B, C, D, inputs='u', outputs='x', states='x', name='plant')

murrayrm · 2023-07-02T16:17:30Z

examples/mrac_siso_direct_lya_rule_statespace.py

+G_model_ss = ct.StateSpace(Am,Bm,Cm,Dm)
+
+io_ref_model = ct.LinearIOSystem(
+    G_model_ss,
+    inputs=('r'),
+    outputs=('xm'),
+    states=('xm'),
+    name='ref_model'
+)


As above, use ss instead.

murrayrm · 2023-07-02T16:18:23Z

examples/mrac_siso_direct_lya_rule_statespace.py

+)
+
+# Overall closed loop system
+io_closed = ct.InterconnectedSystem(


Use ct.interconnect rather than calling the class constructor.

murrayrm · 2023-07-02T16:18:54Z

examples/mrac_siso_direct_mit_rule_statespace.py

+G_plant_ss = ct.StateSpace(A,B,C,D)
+
+io_plant = ct.LinearIOSystem(
+    G_plant_ss,
+    inputs=('u'),
+    outputs=('x'),
+    states=('x'),
+    name='plant'
+)


Replace with call toct.ss.

murrayrm · 2023-07-02T16:19:04Z

examples/mrac_siso_direct_mit_rule_statespace.py

+G_model_ss = ct.StateSpace(Am,Bm,Cm,Dm)
+
+io_ref_model = ct.LinearIOSystem(
+    G_model_ss,
+    inputs=('r'),
+    outputs=('xm'),
+    states=('xm'),
+    name='ref_model'
+)


Replace with call toct.ss.

murrayrm · 2023-07-02T16:19:21Z

examples/mrac_siso_direct_mit_rule_statespace.py

+
+params={"gam":1, "Am":Am, "Bm":Bm, "signb":np.sign(B)}
+
+io_controller = ct.NonlinearIOSystem(


Replace with call toct.nlsys.

murrayrm · 2023-07-02T16:19:31Z

examples/mrac_siso_direct_mit_rule_statespace.py

+)
+
+# Overall closed loop system
+io_closed = ct.InterconnectedSystem(


Replace with call toct.interconnect.

KybernetikJo · 2023-07-03T17:10:04Z

Thanks for the review. I will incorporate your suggestions.

KybernetikJo · 2023-07-04T00:14:40Z

I need help.

In fact, I'm not sure about the recent commits of this PR. I think I did something wrong. The old files have been merged back into the branch "mrac-examples" by "git rebase main".

In steps 8 of the manual, (https://github.com/python-control/python-control/wiki/How-to-contribute-with-a-pull-request)
which branch must be updated & synchronized, main or feature?

Current situation of "mrac-examples".

Files and documentation should be ok.
But many of my commits are suboptimal. They pollute the log history.

Not sure how to fix this.

bnavigator · 2023-07-04T18:39:40Z

doc/examples.rst

+<<<<<<< HEAD
+   mrac_siso_mit
+   mrac_siso_lyapunov
+=======
   mrac_siso_direct_mit_rule_statespace
   mrac_siso_direct_lya_rule_statespace
+>>>>>>> refs/remotes/origin/mrac-examples


This is an artifact of a merge, which must be fixed

Oops I was looking at an intermediate commit. Please rebase so that it does not appear in the commit history.

In reference to your comment: Do a git rebase -i main and "squash" some commits.

Thx, I will try git rebase -i main.

In steps 8 of the manual, (https://github.com/python-control/python-control/wiki/How-to-contribute-with-a-pull-request)
which branch must be updated & synchronized, main or feature?

Both. First get your local main branch in sync with upstream and then rebase feature onto your synced local main

…ys, interconnect)

murrayrm requested changes Jul 2, 2023

View reviewed changes

bnavigator reviewed Jul 4, 2023

View reviewed changes

Add direct mrac siso examples, using new input/output system (ss, nls…

f3713f1

…ys, interconnect)

KybernetikJo force-pushed the mrac-examples branch from 0c04a2c to f3713f1 Compare July 5, 2023 18:27

KybernetikJo requested a review from murrayrm July 5, 2023 18:59

murrayrm merged commit a8658e4 into python-control:main Jul 7, 2023

murrayrm added this to the 0.10.0 milestone Mar 31, 2024

		mrac_siso_direct_mit_rule_statespace
		mrac_siso_direct_lya_rule_statespace

		@@ -0,0 +1,15 @@
		Model-Reference Adaptive Control (MRAC) siso, direct lyapunov rule

		@@ -0,0 +1,185 @@
		# mrac_siso_direct_lya_rule_statespace.py
		# June 2023


		params={"gam":1, "Am":Am, "Bm":Bm, "signb":np.sign(B)}

		io_controller = ct.NonlinearIOSystem(

Comments

Conversation

KybernetikJo commented Jun 15, 2023

Uh oh!

murrayrm commented Jun 19, 2023

Uh oh!

coveralls commented Jun 19, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

murrayrm left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

KybernetikJo commented Jul 3, 2023

Uh oh!

KybernetikJo commented Jul 4, 2023

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

coveralls commented Jun 19, 2023 •

edited

Loading