Murray Wiki - User contributions [en]

Art Krener, Mar 2014

2014-03-03T16:14:06Z

Mhorowit: /* Schedule */

Art Krener will be visiting Caltech on 3 Mar 2014 (Mon). IF you would like to meet with him, sign up below.

=== Schedule ===
* 10:30 - Richard Murray, 109 Steele Lab
* 11:00 - Seminar, 121 Annenberg
* 12:00 - Lunch with Richard and John
* 1:30 - Eric Wolff, 331 Annenberg
* 2:15 - Matanya Horowitz, 335 Annenberg
* 3:00 - Yilin Mo, 310 Annenberg
* 3:45 - Done for the day

=== Abstract ===

Computational Issues in Nonlinear Control

Arthur J. Krener, Naval Postgraduate School 
3 March 2014 - 11:00am PST, 121 Annenberg

Over the past several decades there has been tremendous progress in the development of nonlinear systems theory but implementation of these ideas have lagged behind because of the lack of effective computational tools. Frequently the nonlinear theory requires the solution of partial differential equations in continuous time and functional equations in discrete time. Algorithms for these have been developed but further work is needed to make them broadly and easily accessible. We believe that computational nonlinear control is in a similar stage of development that computational linear control was in around early 1980s. Then there was a well developed theory of linear control but computational tools lagged behind. Soon after comprehensive tools such as MATLAB and Matrix X were developed and put to great use in implementing the linear theory. Advancements in numerical methods together with the exponential increase in computational power has made it possible to solve complex nonlinear problems, many of which are closely related to control systems applications. Developing computational algorithms and software tools for such control systems are not only promising, but also necessary.

The three topics that we will focus on are the following.
# Numerical solution of Hamillton-Jacobi-Bellman and Dynamic Programming equations
# Numerical calculation of optimal trajectories
# Numerical calculation of invariant manifolds

Art Krener, Mar 2014

2014-02-25T05:21:44Z

Mhorowit: /* Schedule */

Art Krener will be visiting Caltech on 3 Mar 2014 (Mon). IF you would like to meet with him, sign up below.

=== Schedule ===
* 10:30 - Richard Murray, 109 Steele Lab
* 11:00 - Seminar, 121 Annenberg
* 12:00 - Lunch with Richard, John?, Doug?
* 1:30 - Matanya Horowitz, 335 Annenberg
* 2:15 - Open
* 3:00 - Open
* 3:45 - Open
* 4:30 - Done for the day

=== Abstract ===

Computational Issues in Nonlinear Control

Arthur J. Krener, Naval Postgraduate School 
3 March 2014 - 11:00am PST, 121 Annenberg

Over the past several decades there has been tremendous progress in the development of nonlinear systems theory but implementation of these ideas have lagged behind because of the lack of effective computational tools. Frequently the nonlinear theory requires the solution of partial differential equations in continuous time and functional equations in discrete time. Algorithms for these have been developed but further work is needed to make them broadly and easily accessible. We believe that computational nonlinear control is in a similar stage of development that computational linear control was in around early 1980s. Then there was a well developed theory of linear control but computational tools lagged behind. Soon after comprehensive tools such as MATLAB and Matrix X were developed and put to great use in implementing the linear theory. Advancements in numerical methods together with the exponential increase in computational power has made it possible to solve complex nonlinear problems, many of which are closely related to control systems applications. Developing computational algorithms and software tools for such control systems are not only promising, but also necessary.

The three topics that we will focus on are the following.
# Numerical solution of Hamillton-Jacobi-Bellman and Dynamic Programming equations
# Numerical calculation of optimal trajectories
# Numerical calculation of invariant manifolds

Dennis Bernstein, Jan 2014

2014-01-29T23:39:12Z

Mhorowit: /* Schedule */

Dennis Berstein and James Forbes from U. Michigan will be visiting Caltech on 29 January 2014. If you would like to meet with Dennis and Jim during their visit, please sign up below.

=== Schedule ===
* 12:30 - Lunch and meeting with Richard
* 2:00 - Seminar, 121 Annenberg
* 3:00 - CDS tea
* 3:30 - Doug MacMartin
* 4:00 - Yilin Mo
* 4:30 - Matanya Horowitz, 335 Annenberg
* 5:00 - Eric Wolff, 331 Annenberg
* 5:30 - Done for the day

=== Abstract ===
<center>
How Much Modeling Information Is Really Needed for Feedback Control?

Dennis Bernstein 
University of Michigan
</center>

Modeling for control is often expensive and time-consuming—not to mention futile, especially when a plant changes unpredictably. Our research is therefore aimed at the following fundamental question: What is the minimal modeling information (order, parameters, nonlinearities, noise spectra, etc.) that must be known—and how *well* must it be known—so that a controller can reliably meet performance specifications?
The approach we are developing is based on retrospective cost adaptive control (RCAC), which uses retrospective optimization for online learning. RCAC is easy to implement, and requires extremely limited modeling information. In this talk I will explain the rationale for RCAC, its applicability to various types of plants (stable/unstable, minimum-phase/NMP, SISO/MIMO, linear/nonlinear), the modeling information it can operate with and (especially) without, and the status of its theoretical foundation.
For flight control, we will apply RCAC to the extreme case of totally unknown control-surface faults, such as a stuck rudder or severe rate saturation. Additional examples are taken from missile control, noise and vibration control, and spacecraft attitude control with nonlinear actuation such as CMGs.

Dennis Bernstein, Jan 2014

2014-01-25T22:39:15Z

Mhorowit: /* Schedule */

Dennis Berstein and James Forbes from U. Michigan will be visiting Caltech on 29 January 2014. If you would like to meet with Dennis and Jim during their visit, please sign up below.

=== Schedule ===
* 12:30 - Lunch and meeting with Richard
* 2:00 - Seminar, 121 Annenberg
* 3:00 - CDS tea
* 3:30 - Open
* 4:00 - Open
* 4:30 - Matanya Horowitz
* 5:00 - Open
* 5:30 - Done for the day

=== Abstract ===
<center>
How Much Modeling Information Is Really Needed for Feedback Control?

Dennis Bernstein 
University of Michigan
</center>

Modeling for control is often expensive and time-consuming—not to mention futile, especially when a plant changes unpredictably. Our research is therefore aimed at the following fundamental question: What is the minimal modeling information (order, parameters, nonlinearities, noise spectra, etc.) that must be known—and how *well* must it be known—so that a controller can reliably meet performance specifications?
The approach we are developing is based on retrospective cost adaptive control (RCAC), which uses retrospective optimization for online learning. RCAC is easy to implement, and requires extremely limited modeling information. In this talk I will explain the rationale for RCAC, its applicability to various types of plants (stable/unstable, minimum-phase/NMP, SISO/MIMO, linear/nonlinear), the modeling information it can operate with and (especially) without, and the status of its theoretical foundation.
For flight control, we will apply RCAC to the extreme case of totally unknown control-surface faults, such as a stuck rudder or severe rate saturation. Additional examples are taken from missile control, noise and vibration control, and spacecraft attitude control with nonlinear actuation such as CMGs.

Group Schedule, Fall 2012

2012-11-08T06:46:54Z

Mhorowit: /* Week 7: 12-16 Nov */

This page contains information about various upcoming events that are of interest to the group. __NOTOC__
{| width=60%
|- valign=top
| width=50% |
* [[Schedule|Richard's calendar (travel)]]
| width=50% |
* [[Group Schedule, Summer 2012]]
|}

The schedule for group and subgroup meetings is given below. Everyone should sign up for times to talk in the subgroup meetings. Unless otherwise noted, here are the locations of the meetings:
:{| width=100%
|- valign=top
| width=30% |
* Group meetings - 114 Steele Lab
| width=30% |
* Biocircuits subgroup - 111 Keck
| width=30% |
* NCS subgroup - 110 Steele
|}

{| width=100% border=1
|- valign=top
| width=30% |
=== Week 1: 1-5 Oct ===
'''Biocircuits: 4 Oct (Thu), 10 am - 12 pm'''
* Gita Mahmoudabadi (30 min)
* Victoria (15 min)
* Zach (15 min)
* Lab updates

'''NCS: 4 Oct (Thu), 4-5:30 pm'''
* Necmiye (15 min)
* Research update (15 min)
* Git tutorial by Scott (15 min)
* Project updates

<hr>
* Richard out of town Tue-Wed
* No group meeting this week

| width=30% |

=== Week 2: 8-12 Oct ===
'''NCS: 10 Oct (Wed), 4:30-6:00 pm'''
* <strike> Andrea </strike> Adam (<s>15</s> 45 min)
* Marcella (<s>15</s> 45 min)
* <strike> Mumu </strike> (15 min)
* Project updates

'''Biocircuits: 11 Oct (Thu), 10 am - 12 pm'''
* Nadine (30 min)
* Research update (15 min)
* Anu (15 min)
* Lab updates

<hr>
* Richard out of town Mon, Tue am, Fri pm
* Aaron Ames visiting on Thu (seminar @ noon)
* No group meeting this week
| width=30% |

=== Week 3: 15-19 Oct ===
* Richard out of town all week
* Biocircuits lab cleanup: 16 Oct (Tue), 10 am - 12 pm

|- valign=top
| width=30% |
=== Week 4: 22-26 Oct ===
'''Biocircuits: 23 Oct (Tue), 10 am - 12 pm'''
* Enoch (30 min)
* Jongmin (15 min)
* Vanessa (15 min)
* Lab updates

'''NCS: 23 Oct (Tue), 4-5:30 pm'''
* Eric (15 min)
* <s>Scott Livingston (15 min)</s>
* Mumu (15 min)
* Project updates

<hr>
* Ram Vasuvian visiting on Mon (seminar @ 11)
* DARPA Living Foundries west coast meeting on Thu (Richard out)
* No group meeting this week

| width=30% |

=== Week 5: 29 Oct - 2 Nov ===
'''Biocircuits: 30 Oct (Tue), 10 am - 12 pm'''
* Marcella (30 min)
* <s>Anandh</s> (15 min)
* dan (15 min)
* Lab updates

'''NCS: 31 Oct (Wed), 4:30-6:00 pm'''
* Shuo (15 min)
* Scott? (15 min)
* <s>Research update (15 min)</s>
* Project updates

<hr>
* Richard out of town on Tue
* iGEM World Jamboree, Fri-Mon
* No group meeting this week
| width=30% |

=== Week 6: 5-9 Nov ===
'''Biocircuits: 6 Nov (Tue), 10 am - 12 pm'''
* Anu (30 min)
* Emzo (15 min)
* Nadine (15 min)
* Lab updates

<s>'''Group meeting: 6 Nov (Tue), 12-1:15 pm'''</s>
* To be rescheduled (Katrina Ligett)

'''NCS: 7 Nov (Wed), 4:30-6:00 pm'''
* Fei (15 min)
* Necmiye (15 min)
* Ivan (15 min)
* Project updates

|- valign=top
| width=30% |

=== Week 7: 12-16 Nov ===
'''Biocircuits: 15 Nov (Thu), 10 am - 12 pm'''
* Ophelia (30 min)
* Zach (15 min)
* Enoch (15 min)
* Anandh (15 min)
* Lab updates

'''Group meeting: 15 Nov (Thu), 12-1:15'''
* Wang Sang Koon

'''NCS: 15 Nov (Thu), 4-5:30 pm'''
* Matanya (15 min)
* Mumu (15 min)
* Quentin (15 min)

<hr>

| width=30% |

=== Week 8: 19-23 Nov ===
* Richard out of town all week
* Biocircuits lab cleanup: 20 Nov (Tue), 10 am - 12 pm
* Thanksgiving break, Thu-Fri
* No group meeting this week

| width=30% |
=== Week 9: 26-30 Nov ===
'''Biocircuits: 27 Nov (Tue), 10 am - 12 pm'''
* Vanessa (30 min)
* Jongmin (15 min)
* Victoria (15 min)
* Ishan (15 min)
* Lab updates

'''NCS: 28 Nov (Wed), 4:30-6:00 pm'''
* Eric (15 min)
* Scott Livingston (15 min)
* Marcella (15 min)
* Project updates

<hr>
* DARPA breadboards telecon, Thu @ 9:30 am
* SoCal Control Workshop, Fri (UC San Diego)
* No group meeting this week

|- valign=top
| width=30% |

=== Week 10: 3-7 Dec ===
'''NCS: 5 Dec (Wed), 4:30-6:00 pm'''
* Shuo (15 min)
* Anandh (15 min)
* Stephanie (15 min)
* Project updates

'''Biocircuits: 6 Dec (Thu), 10 am - 12 pm'''
* Emzo (30 min)
* Marcella (15 min)
* Dan (15 min)
* Lab updates

'''Group meeting: 6 Dec (Thu), 12:00-1:15 pm"
* German Enciso (UC Irvine)

<hr>

| width=30% |

=== Week 11: 10-14 Dec ===
* CDC week, no group meetings
* Biocircuits lab cleanup: 11 Dec (Tue), 10 am - 12 pm
<hr>
* Richard out, Mon-Thu

| width=30% |

=== Week 12: 17-21 Dec ===
* Winter break, no group meetings
* Richard around Mon-Thu
|}

Aaron Ames, October 2012

2012-10-08T18:23:36Z

Mhorowit: /* Friday (12 Oct) */

Aaron Ames will be visiting Caltech on 11-12 October (Thu-Fri). If you would like to meet with him, sign up below.

{| width=100% border=1
|- valign=top
| width=50% |
==== Thursday (11 Oct) ====
* 12:00: CDS seminar (114 Steele Lab)
* 1:15: Richard
* 2:30: Open
* 3:15: Open
* 4:00: Necmiye
* 4:45: Open
|

==== Friday (12 Oct) ====
* 9:45: Open
* 10:30: Matanya Horowitz
* 11:15: Open
* 12:00: Lunch meeting with John's group
* 2:00: Additional discussions (TBD)
|}

=== Seminar Abstract ===

<center>
'''Simplicity on the Far Side of Complexity in the Control of Bipedal Robots'''

Speaker: Aaron Ames 
Affiliation: ME, Texas A&M 
Date and time: 11 October 2012 - 12:00pm 
Location: 114 Steele Lab
</center>

Humans have the ability to walk with deceptive ease, navigating everything from daily environments to uneven and uncertain terrain with efficiency and robustness. Despite the simplicity with which humans appear to ambulate, locomotion is inherently complex due to highly nonlinear dynamics and forcing. Yet there is evidence to suggest that humans utilize a hierarchical subdivision between cortical control and central pattern generators in the spinal column, indicating that when humans perform motion primitives potentially simple and characterizable control strategies are implemented, i.e., humans display simplicity on the far side of complexity. If these fundamental mechanisms underlying human walking can be discovered and formally understood, human-like abilities can be imbued into robotic devices with far-reaching applications ranging from legged robots for space exploration to disaster response.

This talk presents the process of formally achieving bipedal robotic walking through controller synthesis inspired by human locomotion, and demonstrates these methods through experimental realization on multiple bipedal robots. Motivated by the hierarchical control present in humans, we claim that the essential information needed to understand walking is encoded by a simple class of functions canonical to human walking. In other words, we view the human as a complex system, or "black box," and outputs of this system (as computed from human locomotion data) are presented that appear to characterize its behavior—thus yielding low dimensional characterization of human walking. By considering the equivalent outputs for the bipedal robot, a nonlinear controller can be constructed that drives the outputs of the robot to the output of the human; moreover, the parameters of this controller can be optimized so that stable robotic walking is provably achieved while simultaneously producing outputs of the robot that are as close as possible to those of a human. The end result is the automatic generation of bipedal robotic walking that is remarkably human-like and is experimentally realizable, as will be evidenced by the demonstration of the resulting controllers on multiple robotic platforms.

Biography: Dr. Aaron D. Ames is an Assistant Professor in Mechanical Engineering at Texas A&M University, with a joint appointment in Electrical and Computer Engineering. His research interests center on robotics, nonlinear control, and hybrid and cyber-physical systems, with special emphasis on bipedal robots, behavior unique to hybrid systems such as Zeno behavior, and the mathematical foundations of hybrid systems. Dr. Ames received a BS in Mechanical Engineering and a BA in Mathematics from the University of St. Thomas in 2001, and he received a MA in Mathematics and a PhD in Electrical Engineering and Computer Sciences from UC Berkeley in 2006. At UC Berkeley, he was the recipient of the 2005 Leon O. Chua Award for achievement in nonlinear science and the 2006 Bernard Friedman Memorial Prize in Applied Mathematics. Dr. Ames served as a Postdoctoral Scholar in the Control and Dynamical System Department at the California Institute of Technology from 2006 to 2008. In 2010 he received the NSF CAREER award for his research on hybrid systems and bipedal robotic walking.

Henrik Sandberg, July 2012

2012-07-25T05:09:26Z

Mhorowit: /* Monday */

Henrik Sandberg will be visiting CDS on 30-31 July (Tue-Wed). __NOTOC__

=== Schedule ===
{| width=100% border=1
|- valign=top
| width=50% |
==== Monday ====
* 10 am: Open
* 11 am: Seminar, location TBD
* 12 noon: Lunch
<hr>
* 1:30 pm: Matanya Horowitz
* 2:15 pm: Open
* 3:00 pm: Open
* 3:45 pm: Open
* 4:30 pm: Open
|

==== Tuesday ====
* 10 am: Enoch
* 10:45 am: Open
* 11:30 am: Open
<hr>
* 1:30 pm: Open
* 2:15 pm: Open
* 3:00 pm: Open
* 3:45 pm: Open
* 4:30 pm: Open
<hr>
* 6:30 pm: Richard
|}

=== Abstract ===

TBD