Welcome to the CATS!

navigation icon Research

navigation icon Facilities

navigation icon Faculty & Staff

navigation icon News

navigation icon Programs

navigation icon Contact CATS

navigation icon Affiliates

navigation icon Seminars and Events

navigation icon Forms and Documents

navigation icon Home

nystar logo

Rensselaer logo

CATS Seminars – Past Seminars, Spring 2006

navigation icon Current Seminar Schedule
Past Seminars
navigation icon Fall 2005
navigation icon Springl 2006
navigation icon Fall 2006


CATS Seminar, November 21, 2006, 10:00am-11:00am, CII 7003

Precision Control of Micro- and Nano-Scale Systems

Jason J. Gorman, Ph.D.
Intelligent Systems Division, Manufacturing Engineering Laboratory, National Institute of Standards and Technology (NIST)

Similar to macro-scale machines and systems, the application of control systems at the micro- and nano-scales is critical in achieving precise operation. However, the shortened length and time scales; the presence of mechanical and thermal noise; and the prevalence of strong nonlinearities, among many other issues, present many new challenges not typically found at the macro-scale. This seminar will discuss many of these issues, and will present two specific examples currently being explored within the nanomanufacturing program at NIST. The first example is a class of micro-scale nanopositioning mechanisms that are being developed for applications in nanomanipulation and nano-scale metrology. These mechanisms must perform with the same nano-scale positioning resolution and accuracy as their macro-scale counterparts. Open- and closed-loop precision motion control approaches have been designed for a prototype mechanism, which balance performance and robustness with ease of implementation. The second example is the control of optically trapped particles for precision nanoassembly of nanoparticles, and the characterization of cellular interactions with nanostructures. Optical trapping is a method for manipulating cells, macromolecules, and particles with sizes ranging from 10 micrometers down to 25 nanometers. Brownian motion of the trapped particle typically limits assembly precision and inhibits trapping lifetime for particles below 100 nm. A closed loop control approach has been developed to suppress Brownian motion and maximize trapping strength. Theoretical and experimental results for both of these examples will be presented, and their relevance to general approaches for controlling micro- and nano-scale systems will be discussed.

Jason J. Gorman received a B.S. in aerospace engineering from Boston University in 1994, and an M.S. and Ph.D. in mechanical engineering from The Pennsylvania State University, in 2000 and 2002, respectively. In 1997, he was awarded a National Science Foundation Graduate Research Traineeship. Upon graduating, he was awarded a National Research Council Postdoctoral Research Associateship at the National Institute of Standards and Technology (NIST). Currently, he is a research associate in the Manufacturing Engineering Laboratory at NIST, where he works within the nanomanufacturing program. His research interests include nanomanipulation; MEMS/NEMS; robust and nonlinear control; and nanoinstrumentation.

Refreshments served

Return to top

CATS Seminar, November 10, 2006, 1:00pm-2:00pm, JEC 3117

Aggressive Maneuvering Helicopter Flight Control using Robust Hybrid Automata

Leena Singh, Charles Stark Draper Laboratories

Traditional control algorithms impose performance limitations that fall far short of what can be achieved by skilled human pilots. The main objective of this research was to develop and demonstrate the ability for aggressive, real-time, highly-maneuverable trajectory generation and stabilizing, high-bandwidth tracking control. The goal of the project under the DARPA-SEC program was to demonstrate autonomous vehicle maneuvering capability well beyond the level typical in today's autonomous air vehicles. The DARPA-SEC demonstrator platform for in-flight demonstrations of the algorithms was Georgia Tech's Yamaha RMAX helicopter - the GTMAX.

The robust hybrid automata (RHA) guidance algorithm is based on two key concepts: (1) output feedback linearization that decomposes the nonlinear dynamics state space into exterior and interior state subspaces and (2) recognizing that the resulting model is differentially flat. The exterior state dynamics are a function of the exterior and interior state dynamics and compose the output space. The trajectory planning (guidance) is done in this exterior subspace. The interior state dynamics are an invertible function of the control and are independent of the exterior states. Given this separation of plant dynamics into exterior and interior dynamics, we synthesize trajectories (in the interior subspace) from a collection of trim and maneuver elements. Trim conditions lie on the equilibrium manifold of the internal dynamics (ddt(xint) = 0) which is sampled to compose a trim library. Additionally, we introduce a maneuver library - these present stable, admissible transitions between trim states. These maneuvers are computed and stored in the input space by utilizing the invertibility of the differentially flat dynamic structure.

Trajectory planning involves stitching together the discrete trim and maneuver elements to connect points in a Cartesian planning space with a realizable trajectory. Dynamic planning with a pre-computed cost-to-go table is parameterized in a (Cartesian + Trim) space. Agile, highly maneuverable trajectories are synthesized and command an inner loop with trajectory set-points and associated feedforward commands necessary to follow that trajectory. We have demonstrated this algorithm in flight on Draper's X-cell helicopters and SEC's GTMAX demonstrator. With the GTMAX, flight demonstrations involved flying the periphery of the McKenna MOUT site in Ft. Benning, GA at building-top level and conducting rapid descent to beneath building top level to evade a "sniper". Simulations will be shown at the briefing.

Leena Singh received a BS in Physics/Math from Mt. Holyoke College and MS & PhD in Electrical Engineering from Rensselaer. Dr. Singh worked at United Technologies Research Center for 4 years as a Control Systems engineer where she worked on the Fire Control System for the Comanche Helicopter (Boeing, Sikorsky project), advanced concepts to reduce wake shocks that cause premature high-cycle fatigue failure in aircraft engine rotor blades (for Pratt and Whitney), and hybrid building control systems with Carrier, Otis and Pratt-Whitney. Dr. Singh has been at Draper Laboratory since 2001 designing and flight-testing aggressive maneuvering, high-performance guidance and control algorithms for UAVs (helicopters), large and small scale parafoil systems and attitude and formation flight of micro-satellite systems. More recently, she worked on the flight control system for NASA's Crew Exploration Vehicle concepts for the CEV competition.

Refreshments served

Return to top

CATS Seminar, November 3, 2006, 1:00pm-2:00pm, JEC 3117

MEMS and Microfluidics at The Frontiers of Science and Technology

Anis Zribi, General Electric Global Research Center

The miniaturization of devices and instruments is a continuing trend that finds its roots in miniaturizing mechanical, optical and most recently electronic devices. The driving forces for device miniaturization are numerous and keep increasing as we encounter and develop new applications. Device cost and portability are major incentives for miniaturization but additional reasons include higher performance, integration of multiple functions, reduction of occupied space, lower device-to-device variability…etc. Micro devices and instruments are becoming key enablers for a broad range of industrial, medical, military and security applications. Micro sensors, for example, are used to optimize the performance of a system, monitor its health or the surrounding operating conditions in its environment. Micro sensor data can be as simple as pressure, temperature, moisture, chemical and biological analyte concentrations, vibration, speed, or as complex as a spectrum. Device and instrument miniaturization are rapidly growing research areas at the Global Research Center of GE where advanced research for all of the GE businesses is conducted. This talk will present an overview of the growth of micro and nano enabled sensors research and the key drivers of miniaturization. The speaker will highlight the research and development challenges and provide examples of current MEMS-based sensor research programs ongoing at GE Research.

Anis Zribi is an Applied Physicist at the Global Research Center (GRC) of General Electric, where he is leading research in the area of microsystems and microfluidics for chemical and biological analyses. He received an M.S.E. in Physics from the Polytechnic School of Engineering (1996 Paris & Grenoble, France), an M.Sc. in Materials Physics from Chalmers University of Technology (1998 Gothenburg, Sweden) and a Ph.D. in Materials Science from the State University of New York in Binghamton (2002). Since joining GRC in 2002, Dr. Zribi has led and contributed to several projects including the Nanotechnology Advanced Technology Program, the Photonics Advanced Technology Program and a few MEMS sensors and actuators projects. His research interests and activities at GRC include MEMS for micro and nano enabled analytical instruments (spectrometers), gas sensors and nanomaterials characterization tools. Dr. Zribi holds 5 patents and over 25 pending patent applications in MEMS, photonics and sensors.

Refreshments served

Return to top

CATS Seminar, October 27, 2006, 1:00pm-2:00pm, JEC 5030

Cooperative Control of Collective Motion and Ocean Sampling with Autonomous Vehicles

Derek Paley, Princeton University

Collective motion is a ubiquitous phenomenon in nature and an essential aspect of emerging engineering applications such as ocean sensor networks. The study of mechanisms that link what individuals do to what a group does is a topic with significant consequences in ecology and evolutionary biology. Cooperative control of underwater vehicles enables scientists to make new discoveries that affect our understanding of the environment. This talk will describe a feedback control framework for stabilization of relative equilibria in a model of identical particles moving in the plane at unit speed. Relative equilibria correspond to either parallel motion of all particles with fixed relative spacing or circular motion of all particles around the same circle. Particles exchange information according to a communication graph that can be undirected or directed and time- invariant or time-varying. The talk will describe extensions to the framework that are of particular relevance to sensor network applications. Lastly, recent experimental results from the demonstration of cooperative control of a fleet of autonomous underwater vehicles will be presented.

Derek A. Paley is a Ph.D. candidate in the Department of Mechanical and Aerospace Engineering of Princeton University. He received the B.S. degree in applied physics from Yale University in 1997. From 1997 to 2000, he worked as an analyst in the defense industry for Metron Inc. From 2002 to 2002, he worked as a software engineer in the underwater vehicle industry for Bluefin Robotics Corp. He received the M.A. degree in mechanical and aerospace engineering from Princeton University in 2004. His research interests are collective motion of natural and engineered systems, cooperative control, and ocean sampling with autonomous vehicles.

Refreshments served

Return to top

CATS Seminar, September 1, 2006, 1:00pm-2:00pm, JEC 3117

Finding Structured Locomotion Error Recovery Modes

Ravi Balasubramanian, Robotics Institute, Carnegie Mellon University

A robot's locomotion mode fails when its environmental contacts fail, a situation called a locomotion error.  For example, a legged robot cannot move when its leg becomes trapped in a crevice, and a wheeled robot is handicapped when its wheels skid.  How can a robot recover when its standard locomotion mode fails?  One way is to utilize any remaining freedoms to move the robot to a situation where the robot's standard locomotion mode is again feasible.  However, planning and controlling such unconventional motion is difficult, since the relationship between the robot's controls and its motion in a locomotion error is unclear and there is significant uncertainty.  My research proposes finding recovery strategies by exploiting the structure inherent to the robot's constrained mobility and environmental interaction in the locomotion error.  A robot equipped with multiple locomotion modes can choose between them depending on the circumstances, ultimately achieving robust mobility.

While robotic locomotion fails in many ways depending on the robot's design and the environmental interaction, my research finds novel recovery modes involving a combination of direct actuation and dynamically coupled actuation for two specific locomotion errors: first, a high-centered legged robot, where the robot's body is stuck on a rock and the robot's legs dangle in air; and second, a car trapped in a slippery pit.

Ravi Balasubramanian recently received his PhD in Robotics from Carnegie Mellon University, working with Prof. Matthew T. Mason and Prof. Alfred A. Rizzi.  His core research area is modeling, control, and planning for mechanical systems, with experience gained from his thesis research on novel control strategies for mobile-robot error recovery, his internships at the University of Michigan (legged locomotion group) and Anybots Inc. (teleoperated humanoid manipulation), and his Mechanical Engineering education from the National University of Singapore.  Ravi's thesis research was a finalist in the best student paper competition at the International Conference on Robotics and Automation, 2004 (top 3 from a total 1000 papers).  Also, Ravi has co-taught (with Dr. George Kantor) a section of a robotic manipulation course in Carnegie Mellon, designing the curriculum and guiding course projects.  His research website is at http://www.cs.cmu.edu/~bravi/research.html.

Refreshments served

Return to top

9/1/06: Ravi Balasubramanian (Carnegie Mellon University)

9/11/06: R. Scott Erwin (Air Force Research Lab)

9/14/06: Michael Wang (Chinese University of Hong Kong)

10/19/06: Sanjay Joshi (University of California at Davis)

10/20/06: Manish Sinha (General Motors)

10/27/06: Derek A. Paley (Princeton Univeresity)

11/3/06: Anis Zribi (General Electric Global Research)

11/10/06: Leena Singh (Draper Lab)

11/10/06: Jason Gorman (NIST)

Joint CATS/MANE/CS Seminar, September 14, 2006, 11:00am-12:00pm, CII 5003

Level Set Method Based Design of Compliant Mechanisms for Passive Self-Alignment of Hybrid MEMS Assembly

Michael Y. Wang, Department of Automation & Computer-Aided Eng., Chinese University of Hong Kong

A monolithic compliant mechanism transmits applied forces from specified input ports to output ports by elastic deformation of its comprising materials, fulfilling required functions analogous to a rigid-body mechanism. It has a large range of applications in both micro and macro domains. This presentation describes a level-set method for designing monolithic mechanisms with distributed compliance and/or made of multiple materials. Central to the method is a level-set model that precisely specifies the distinct material regions and their sharp interfaces as well as the geometric boundary of the structure, capable of performing topological changes and capturing geometric evolutions at the interface and the boundary. Techniques for eliminating de facto hinges and for geometric control in the design are discussed, aiming at producing more reliable compliant mechanism designs for MEMS devices. We further discuss the intrinsic deficiencies in the widely used "spring model" and propose a new
formulation considering the "characteristic stiffness" of the mechanism. The result is a design with highly even-distributed compliance and a more desirable characteristic, which uniquely distinguishes our method. These methods are demonstrated with
benchmark examples of both structure optimization and compliant mechanism optimization. The compliant mechanisms are intended for the use in automated assembly of hybrid MEMS with self-alignment techniques to eliminate tight positioning requirements.

Michael Yu Wang is a Professor at the Chinese University of Hong Kong, after ten years with the Department of Mechanical Engineering, University of Maryland. He has numerous professional honors-National Science Foundation Research Initiation Award, 1993; Ralph R. Teetor Educational Award from Society of Automotive Engineers, 1994; LaRoux K. Gillespie Outstanding Young Manufacturing Engineer Award from Society of Manufacturing Engineers, 1995; Boeing-A.D. Welliver Faculty
Summer Fellow, Boeing, 1998; Chang Jiang (Cheung Kong) Scholars Award from the Ministry of Education of China and Li Ka Shing Foundation (Hong Kong). He received the Kayamori Best Paper Award of 2001 IEEE International Conference on Robotics and Automation (with D. Pelinescu.) He is a Senior Editor of IEEE Trans. on Automation Science and Engineering, and served as an Associate Editor of IEEE Trans. on Robotics and Automation and ASME Journal of Manufacturing Science and Engineering. He is a Distinguished Lecturer of IEEE Robotics and Automation Society (2006-2008). His research interests include computational design and optimization of solids, precision engineering, and electronic and photonic manufacturing, with 190
technical publications in these areas. He received his Ph.D degree from Carnegie Mellon University (1989). He is a Fellow of ASME.

Return to top

CATS Seminar, September 11, 2006, 11:00am-12:00pm, JEC5030

Research & Development for Future Military Space Applications

R. Scott Erwin, Space Vehicle Directorate, Air Force Research Lab

The United States relies on space-based assets for a large variety of military applications, from the ability for precise navigation and precision strike provided by the Global Positioning System constellation of satellites to high-bandwidth secure communications capability to link our national and theatre commanders to the forces in the field that is provided by military communications satellites. But space, once the domain of only a select few countries with sufficient political and financial unity to overcome the natural barriers that govern it, is increasingly becoming available to more and more nations as more countries either develop or purchase the ability to enter space or the products produced by the assets of other space-faring nations.

This change has resulted in a reexamination by the United States of the military use in space, with the result that new missions and capabilities are under consideration for development. These areas include the development of a responsive space capability for gap filling and augmentation missions and the development of a space situational awareness capability that will provide the US military with true understanding of what craft move through and in space and their capabilities.

This talk will focus on the research and development directions that the Air Force Research Laboratory is taking to meet these new challenges, with specific focus on areas that systems and controls as a technical discipline might address in this evolving development effort, and will discuss several areas that the Air Force Research Laboratory's Space Vehicles Directorate has initiated along these lines. The talk will focus on several projects and directions rather than an in-depth technical discussion on any one area of research in order to provide perspective on the subject. The talk will be of benefit to anyone interested in current areas of interest to the Air Force for Space Applications, regardless of their technical discipline.

R. Scott Erwin received a B.S. in Aeronautical Engineering from Rensselaer Polytechnic Institute in 1991, and his M.S. and Ph.D. degrees in Aerospace Engineering from the University of Michigan in 1993 and 1997, respectively. He has been an employee of the Air Force Research Laboratory, Space Vehicles Directorate located at Kirtland AFB, NM, from 1997 to the present. He is currently the Technical Area Lead for Command, Control, and Communications (C3) Research at the Space Vehicles Directorate, where his responsibilities include the fiscal management, planning, and technical direction of research in these areas. Dr. Erwin also has a simultaneous appointment as the Academic Affairs Coordinator for the Space Vehicles Directorate, serving as the Laboratory Center Representative for the NRC Research Fellows Program, the ASEE Summer Faculty Fellows Program, and the AFRL/VS Space Scholars Program, as well as performing general liaison functions with academia. Dr. Erwin is a Senior Member of both the IEEE and the AIAA, and he has served as an Associate Editor for the IEEE Transactions on Control Systems Technology and the IEEE Control Systems Society Conference Editorial Board. Dr. Erwin was selected as the recipient of the 2001 United States Air Force Science and Engineering Award for Exploratory and Advanced Technology Development and was the winner of the IEEE Albuquerque Section's Junior Engineer of the Year Award in 2004. He is a member of external advisory boards for the Electrical and Computer Engineering Department of the University of New Mexico and the Aerospace Engineering and Engineering Mechanics Department of the University of Texas at Austin. He has authored or co-authored over 50 publications including 9 refereed journal papers in the areas of spacecraft dynamics and controls.

Return to top

CATS Seminar, October 19, 2006, 1:00pm-2:00pm, JEC 3117

Following-Control Systems for Autonomous Robots

Sanjay S. Joshi, University of California, Davis

One key functionality for the use of autonomous robots in many application domains is a robot's ability to follow another robot or human being.  Following control can arise in single robot applications, where a robot may follow a leader from one location to another. Or, following control may arise in group-robotics, in which a robot team may follow a single leader or neighbor from one place to another. In this talk, we describe our recent work on two following control methods/applications. In this first study, we discuss the use of adaptive control for a group-robotics task: keeping robots in geometric formation while they are moving.  In the second study, we discuss a human-following problem, in which we use vision-based predictive control and human-specific cues to allow robot following within buildings.

Sanjay S. Joshi is Assistant Professor of Mechanical and Aeronautical Engineering
at the University of California, Davis, where he formed the Robotics, Autonomous
Systems, and Controls Laboratory. He is also a faculty member within the Electrical
Engineering and Computer Science graduate programs. His research interests span
both the theory and application of autonomous systems. Current projects in the laboratory include biorobotics, space exploration, and medical applications. Dr. Joshi graduated with a B.S. from Cornell University in 1990, a M.S. from UCLA in 1992, and a Ph.D. from UCLA in 1996 (all in Electrical Engineering). He then became a Member of the Technical Staff at NASA's Jet Propulsion Laboratory in the Guidance and Control
Analysis Group. From 2000-2001, he was a Visiting Assistant Professor of Engineering at Harvey Mudd College in Claremont, California.

Refreshments served

Return to top

CATS Seminar, October 20, 2006, 1:00pm-2:00pm, JEC 3117

Control of Fuel Cell Systems for Automotive Applications: Challenges and Opportunities

Manish Sinha, General Motors Corporation

Fuel cell power systems, in particular polymer electrolyte membrane (PEM), using hydrogen as fuel have witnessed intense development and offer a clean and efficient alternative for power generation for automotive applications. General Motors (GM) is targeting, and has demonstrated, both hydrocarbon-based (gasoline) as well as direct hydrogen fuel cells. This electrochemical powertrain presents its own set of system challenges including packaging and heat rejection. Control requirements that fuel cell automotive systems must meet, are similar to internal combustion (IC) engines, and include: fast start, fast dynamic operation (0-60mph response), large turndown ratio (ratio of maximum to minimum power), extreme environmental conditions, freeze tolerance, long lifetime and of course, efficiency.

Short drive cycles in typical drive profile and associated heat-up and cool-down can be compared to chemical batch reactors with highly transient throughput demands. Fuel cell system control plays a balancing act in meeting the dynamic power demand from the electric traction system (ETS) reliably while simultaneously ensuring durability of electrochemical “reactor”. Heart of any control system is in understanding the dynamic behavior of the plant through mathematical modeling and exploiting this understanding for improved control and operation. Offline modeling of subsystems also enables rapid prototyping of algorithm concepts. Application of modeling for offline simulations and hardware-in-the-loop (HIL) will be briefly discussed. The sensitivities of process parameters to sensor and actuators dictate the accuracy and response requirements of sensors for fuel cell application. Challenges in the area of process monitoring, sensor and actuator technology will be presented. Moreover, control system provides process monitoring of stack health for fault diagnostic, online remedial action, and adaptive control. To illustrate online process monitoring and diagnostic of fuel cell stack, a pattern based diagnostic will be discussed for early detection of onset of stack instability due to non-uniform cell-to-cell water management.

Manish Sinha is with General Motors Corporation Research and Development in Honeoye Falls, New York.