Research Statement (2002-2007)

Perry Y. Li

Research Statement (2002-2007) Perry Y. Li Department of Mechanical Engineering University of Minnesota [email protected] My research is in the general area of mechatronics, dynamic systems and control. My research program reflects and highlights the applicability of these subjects to a diverse and varied energy domains and fields. My research approach emphasizes exploiting physical insights in a mathematically rigorous manner, and experimental verification of our theories. My main research thrust in the past few years can be categorized into the following groupings with some overlap: 1) Passive machines that interact with humans 2) Innovations in fluid power components and systems 3) Control of other system and machine processes 1. Passive machines that interact with humans A major research focus of mine is in control designs that enable robots and other machinery to operate stably and safely in the presence of humans and other physical environment alike. I pioneer a “passive control” approach in which the closed loop control system appears to be a passive system to its environment. The idea is to exploit the simple physical concept of energy conservation to ensure that the physical coupling between the machine and any passive physical environments is stable. This significantly increases the utility of such machines in safety critical and ill characterized environment, especially when human-machine interactions are involved. Passive decomposition for multiple mechanical systems Many passive machine applications that involve coordination of multiple agents, such as coordinating the master and slave robots in teleoperation, multi-robot coordination and formation flight. We have developed a theory for decomposing and decoupling the system dynamics so that the coordination aspect and the aspect of the system after coordination can be dealt with independently as without violating the passivity property. The theory is developed in a geometric mechanical framework [1,2]. In addition to elucidating the geometric properties of the decoupled systems, it allows passivity preserving controllers to be designed easily. The decomposition approach has been Figure 1. Passive nonlinear teleoperation applied to passive machine applications: bilateral setup. The kidney shape is the allowable teleoperation [3, 4] to achieve the first successful solution for workspace and the arrows correspond to direction of guidance. a passive nonlinear teleoperator system (see below), and robust human power amplifiers [5]. The decomposition theory has also been applied to the problem of multi-spacecraft formation and maneuvering to achieve a decentralized control solution that is more natural and robust solution than other approaches [6, 7].

1

Perry Y. Li Bilateral (electromechanical) teleoperation Our approach to bilateral teleoperation is to enable a human operating a robot remotely to feel as if he and the work environment are manipulating a common passive rigid mechanical tool with programmable passive dynamics and power scalings [3, 4]. Successful development of this approach will benefit telesurgery, flexible automation in manufacturing and the construction industry alike. Using the passive decomposition, we developed the locked system control that corresponds to imparting useful mechanical tool dynamics; and the shape system control that rigidly coordinates the nonlinear master and slave robots. All these can be done while preserving the passivity (with power scaling) of the overall system. In the nonlinear telerobotic system in Fig. 1, useful tool dynamics consists in potential field for obstacle avoidance and passive guidance using passive velocity field approach that I developed earlier [8]. Passivity as applied to hydraulic systems – passified valves and passive teleoperation While the passivity concept is well accepted in the electromechanical domain, its application to hydraulically actuated systems is more recent and is pioneered by our group. A key obstacle to applying the passivity concept to hydraulic systems is that the inherent passivity properties of directional control valves (relative to electric motors) were unknown. Our research group performed the first passivity analysis on valves to show that they are in fact inherently non-passive [9] and proposed passification methodologies, i.e. methods to modify the physical design or to control the valves so that they behave like passive two-port devices, for both single stage [9] and multi-stage valves [10]. The passification algorithm was later generalized [11] using a bond graph representation. Since bond-graphs highlight the physical nature of the system, this new methodology greatly generalizes our previous results and enables a much broader class of mechatronic systems to be passified. The development of these "passive valves" has subsequently enabled the design of the first teleoperated Figure 2. Passive hydraulic teleoperation setup. control of hydraulic systems that possess the passivity property [12, 13]. These control methodologies provide feedback to the humans in a guaranteed stable manner so that he is able to feel the environment that the hydraulic machine is interacting. While the original works in [12, 13] preserve the passivity property, they do not actively compensate for effects of fluid compressibility. At high load situations, this results in coordination error between the master robot and the hydraulic machine that can be improved upon. This issue is resolved by applying the passive decomposition to the system, representing the decomposed system by an active bond graph, and then applying a successive transformation procedure (similar to the bond graph passification procedure) to achieve a set of arbitrary nice dynamics. This results in much improved coordination results without affecting the passivity property as experimentally demonstrated on the hydraulic backhoe setup in Fig. 2 [14]. Passive hydraulic human power amplifier A hydraulic human power amplifier with two assisted degrees of freedom was designed and built by a group of undergraduate students as their capstone project with funding from the NFPA Small Grant Program (Fig. 3). The objective for the machine is to amplify the applied human force in the assisted degrees of freedom so that the human can manipulate heavy objects while being physically connected to the task, similar to the concept of cobots. The paradigm is similar to teleoperation except that the master and slave systems are the same system. Because the machine must simultaneously interact with the human operator and the work environment, maintaining

2

Perry Y. Li passivity (with a power scaling) is important for stable operation. Initially, a force control algorithm was proposed in which the hydraulic actuator’s force is controlled to be a multiple of the human applied force [15]. Although satisfactory force tracking results were obtained during constrained motion, a positive velocity feedback term required by the algorithm is difficult to implement with finite sampling time in free motion. Thus, neglecting the velocity feedback term, while preserves stability, also introduces force tracking during unconstrained free motion. A more elegant algorithm was recently developed in which the velocity feedback term is replaced by the velocity of a Figure 3. Hydraulic Human Power Amplifier. The pitch virtual mechanical system that interacts with the and reach degrees of freedom are hydraulically assisted by actual machine, and the force control task is a hydraulic actuator and a hydraulic motor / rack-andconverted into one of coordinating the velocities pinion . of virtual and actual systems [16]. The reformulation allows the controlled system to possess a bond graph structure making preserving passivity relatively straightforward. We are currently working on incorporating useful passive dynamics such as path guidance or obstacle avoidance to the hydraulic human power amplifiers. Current research Our current research (within the NSF supported ERC for Compact and Efficient Fluid Power) in the area of passive machines that interact of humans focuses on extending our passive teleoperation and passive human power amplification concepts to the pneumatic and chemo-fluidically actuated domains. The reason is that within the ERC, portable power sources for pneumatic and chemofluidic actuation (pressure bottles, portable compressors, or hydrogen peroxide fuel) are being developed. If successful, truly portable, un-tethered, and powerful human assist devices can then become a reality.

2. Innovations in Fluid Power Components and Systems Fluid power and hydraulic systems are used where high power and high force densities are needed. Consequently, they play critical roles in many industry sectors, including automobile, construction, agricultural, manufacturing, aerospace, etc. The Upper Midwest area is a cradle of the U.S. fluid power industry. Fluid power is inherently multi-disciplinary and involves multiple energy domains. It is therefore an ideal subject for highlighting the multi-disciplinary nature of dynamic systems and control. My interest in this area grew out of interactions with local fluid power industries who have expressed needs for research and teaching capabilities in this area. With the financial and equipment support of a consortium of fluid power companies, I established the Fluid Power Control Laboratory, which is used for both teaching and research purposes. Fluid Power activities at the U. of Minnesota continue to expand with the successful application in 2006 for a NSF supported Engineering Research Center (ERC) for Compact and Efficient Fluid Power (http://www.ccefp.org). The center involves seven universities in the U.S.A. I am serving as the Deputy Director responsible for the research program.

3

Perry Y. Li Although fluid power is prevalent, research in fluid power, especially in the U.S., has been rather stagnant. For example, the basic design of hydraulic servo-valves has been unchanged for over 50 years while the use of advanced mechatronics lags a myriad of other applications. My research in fluid power aims to re-visit basic concepts of hydraulic components and systems, and propose new ones in light of new applications, vigorous dynamical systems concepts and recent advances in sensing, actuation and nonlinear control theory, to achieve better performing systems. Valve designs that utilize unstable flow forces In order to alleviate the bandwidth and flow rate limitations of single stage control valves, we developed a class of valves in which we consciously take advantage of fluid flow instability (that conventional valve designs try to avoid). Single stage control valves are inherently cheaper and more reliable than multistage valves. However, as frequency of operation and flow rates increase, single stage valves require large solenoids to overcome the large resistive flow forces. To extend the bandwidth and flow ratings without using very large and expensive solenoids, our research approach exploits both transient and steady fluid flow forces so as to improve the agility of the spool. Transient flow forces are associated with the acceleration of the fluid inside the valve. They increase with spool velocity, and thus appear like damping forces to the spool. It is well known that an aspect of valve geometry, known as the damping length, determines whether positive or negative damping is achieved. Conventionally, positive damping length is chosen to ensure spool stability. In our Figure 4. Valve geometric design is approach, negative damping is chosen to utilize the unstable flow formulated as a robust synthesis forces to assist the spool movement. The spool can subsequently be problem in which the set of design stabilized through feedback control as necessary [17]. Steady flow parameters is the captured as the forces are associated with flow momentum fluxes, increase with controller to be determine, spool displacement, and thus appear as spring forces on the spool. uncertainties in operating conditions are captured as the plant uncertainty. While it is generally believed that steady flow forces at the flow metering orifices are always stable, we have found, through first principle modeling, CFD analysis and extensive experimental validation, a little known fact that fluid viscosity and negative damping lengths can also induce unstable steady force that can improve spool agility significantly [18]. In contrast to transient flow force that is a function of spool velocities so that any unstable flow force is always advantageous, steady flow force is a function of spool displacement. Thus it is important that the unsteady flow force does not over-compensate over a range of operating conditions such as pressure and temperature. Recognizing the similarity with the robust control synthesis problem, the robust valve geometric design problem is cast into one of finding a robust controller in the presence of a set of structured uncertainties. Figure 4 illustrates how a nominal system, design parameters and uncertainties are inter-related in a familiar LFT configuration. This allows us to adapt existing analytical and computational approaches for solving robust control synthesis problems to solve robust design problems such as ours [19]. Based on the experimentally validated flow force models and the robust design methodologies, we have designed and manufactured a prototype valve that utilizes unstable flow forces. It was demonstrated experimentally that it is capable of or exceeds the flow ratings and bandwidths of a commercially size-5 valve using only a set of size-3 solenoids (which are significantly smaller) [20].

4

Perry Y. Li Smart sensing and smart sensors Another thrust of our research in fluid power system is the development of sensing and sensing methodologies. Recognizing that economical means of providing flow and pressure sensing is important for the commercialization of control concepts, we developed, with funding from the industry supported NFPA Cooperate Network of Research (CNR) program, a MEMS based integrated pressure, flow and temperature (PQT) sensor (Fig. 5) that are intended for integration into fluid power components (such as pumps) [21, 22]. Flow measurement is based on pressure difference across a flow bend, thus Figure 5. MEMS integrated PQT sensor minimizing energy loss. In additional to hardware sensors, we have also and a dime. developed soft-sensing (i.e. observer) algorithms for determining, in real time, the valve spool displacement without using costly hardware sensors such as LVDTs. The soft-sensing scheme is unique in that it utilizes readily available solenoid voltage and current information, and the inductance characteristic of the solenoids in a non-linear observer to determine the spool displacement [23, 24]. Current research My current fluid power research is funded by the ERC for Compact and Efficient Fluid and focuses on three areas: 1) On/off valve based control of fluid power systems. The goal of this research is to develop energy efficient means of controlling fluid power systems that minimizes throttling loss. On/off valves can potentially be highly efficient because in both the fully on and fully off states, the throttling loss is small. By switching between these two states, and by varying the duty ratio of the on or off times, the mean flow/pressure can be modulated. We have studied, experimentally and analytically, the effect of variable design and operational parameters on system efficiency [25-26]. The main obstacle to on/off valve based control is the lack of high speed on/off valves. They are necessary to ensure that the time spent during transitions between the two states (when throttling is significant) is minimized. Whereas a conventional valve with a poppet or spool element that travels linearly requires repeatedly accelerating and decelerating the valve spool (so that energy consumption is proportion to the 3rd power of the switching frequency), we are developing a pulse width modulated on/off valve that is based on unidirectional rotary motion. Thus repeated starting and stopping can be avoided and power is only needed to overcome friction (which is proportional to the 2nd power of the switching frequency) [27]1. Moreover, the rotation scavenges energy from the fluid flow so that external energy is not required. This valve has been prototyped and is being integrated into a fixed displacement vane pump to achieve variable displacement function. Our long term goal is to develop methodologies for controlling pump/motors, transformers, hydrostats, and cylinder actuators using on/off valves. 2) Increasing energy storage density A significant deficiency of hydraulic systems is the lack of high energy density storage. Hydraulic energy storage is typically achieved in gas charged accumulators. Compared to electric batteries, their energy densities are 2 orders of magnitude lower. The goal of this high risk/high payoff project is to increase the energy density by an order of magnitude. If successful, it will enable energy regeneration in many mobile applications, such as hydraulic hybrid passenger vehicles. We are currently investigating an alternative storage approach that we called open accumulator [28]2 that can achieve this goal. 1 2

US patent filed US patent filed

5

Perry Y. Li 3) Developing a hydraulic hybrid vehicle. A goal of the ERC for Compact and Efficient Fluid Power is to utilize fluid power to enable significant fuel saving. One of the demonstration test beds is the hydraulic hybrid passenger vehicle with a lofty target gas mileage of 100 mpg. I am leading this effort, and our team has proposed a regenerative hydro-mechanical drive (HMT) for development [29]1. The advantage of the HMT relative to either the parallel and series architecture is that it allows the bulk of the power to be transmitted efficiently through the mechanical transmission. Our proposed HMT architecture also allows for aggressive energy management, independent wheel torque control and regeneration. 3. Control of other system and machine processes Control of imaging and printing process In xerographic color printers, maintaining the consistency and accuracy of the printed relative to the desired one is critical. It is because human eyes are much more sensitive to color variation than to monochrome variation, and the color printing processes are more complex and are more susceptible to environment variations. In this project that is supported by the NSF and Xerox Corporation, our goal is to design control systems that maintain the color accuracy and consistency. The main challenges lie in the limitation in sensing and actuation of xerographic printers. While the number of colors that need to be controlled is near infinite, the color patches used for sensing the printer’s ability to produce a certain color are costly and can only be used sparingly. Similarly, the number of actuators for correcting the print quality is small compared to the number of colors to be controlled simultaneously. To solve this problem, we have proposed the use of time-sequential sampling approach in which the color gamut is sampled using different color patches scheduled at different times. The time varying color gamut can then be reconstructed in real time for feedback control using the time sequential samples [30, 31]. In order to minimize the number of color patches, a provably correct algorithm for optimizing the time-sequential sampling has been developed based Figure 6 Optimal time sequential on lattice theoretic framework [32]. The optimal sequence makes sampling sequence and the spectral trade off between temporal and tonal/color frequencies. Figure 6 packing that results. shows an optimal time sequential tone sampling sequence for a primary color and the resulting packing the repeated spectra. It has been shown that the number of samples can be reduced by 50% (compared to Nyquist) and that the use of time-sequential sampling for feedback has equivalent benefit as the hardware intensive full gamut sampling approach [33]. Paper Manufacturing In a paper machine, water is removed from pulp slurry via successive multiple dewatering devices. The accuracy and uniformity of the moisture content of the paper is key quality measure. Moreover, because of the large energy usage, efficient energy consumption is a major concern. Funded by NSF and TAPPI Foundation (paper industry foundation), Shri Ramaswamy (Dept of Biobased 1

US provisional patent filed

6

Perry Y. Li Products) and I researched a new process control strategy to control moisture content in paper manufacturing that uses vacuum dewatering boxes as actuators and airflow through wet sheet to infer inprocess moisture content measurements. The current approach uses the heater as the main control actuator and an end-of-process sensor for feedback. The proposed approach is potentially more robust, more able to eliminate the effects of disturbances, and utilizes less energy. I am responsible for the development of the new control topology, the algorithm design and analysis; while Prof. Ramaswamy is responsible for developing physically based process and measurement models. With this approach, the uniformity of moisture content control can be improved and the energy cost of paper manufacturing can be reduced. We have developed, for the first time, an experimental set up for investigating vacuum dewatering at actual machine speeds. With this set up, we have developed a model relating air flow permeability to paper moisture, which is needed for surrogate measurement. In the control aspect, we have developed a pre-emptive control strategy [34] (and an adaptive version [35]) that use surrogate measurements from upstream dewatering boxes to compensate for downstream control errors. Because the surrogate measurement model is highly variable, an adaptive calibration algorithm has also been developed that uses an accurate sensor at the end of the process to automatically calibrate the uncertain surrogate measurement models. The adaptive calibration method has been experimentally verified using a setup that mimics the dynamics of vacuum dewatering [36]. Current research We are currently also working on an industry sponsored project to develop high performance control for electrical generators. In particular, various adaptive control algorithms are being pursued. I am also working with colleagues in the medical school to develop a research program on advanced endoscopic surgical devices. References [Most of these papers are available on-line at http://www.me.umn.edu/~pli/publications.html ] [1] D.J. Lee and P.Y. Li, “Passive Decomposition of Multiple Mechanical Systems Under Coordination Requirements,” Proceedings of the IEEE Conference on Decision and Control, Nassau, p1240-1245, December 2005. [2] D. J. Lee and P. Y. Li, “Passive Decomposition of Multiple Mechanical Systems under Coordination Requirements”, IEEE Transactions on Automatic Control (under review) [3] D.J. Lee and P.Y. Li, “Passive Bilateral Feedforward Control of Linear Dynamically Similar Teleoperated Manipulators,” IEEE Transactions on Robotics and Automation. Vol. 19, No. 3, pp. 443456, June 2003. [4] D.J. Lee and P.Y. Li, “Passive Bilateral Control and Tool Dynamics Rendering for Nonlinear Mechanical Teleoperators,” IEEE Transactions on Robotics and Automation. Vol. 21, No. 5, pp. 936951, October 2005. [5] P.Y. Li, “A New Passive Control for Hydraulic Human Power Amplifier,” ASME-IMECE200615056, Chicago, IL, November 2006. [6] D.J. Lee and P.Y. Li, “Passive Decomposition Approach to Formation and Maneuver Control of Multiple Rigid Bodies,” ASME Journal of Dynamic Systems, Measurement and Control. Vol. 129, No.3, September, pp. 662-677, 2007. [7] D.J. Lee* and P.Y. Li, “Formation and Maneuver Control for Multiple Spacecraft,” Proceedings of the 2003 American Control Conference, Denver, CO, June 2003. [8] P.Y. Li and R. Horowitz, “Passive Velocity Field Control of Mechanical Manipulators,” IEEE Transactions on Robotics and Automation. Vol. 15, No. 4, pp. 751-763, 1999.

7

Perry Y. Li [9] P.Y. Li, “Towards Safe and Human Friendly Hydraulics: The Passive Valve,” ASME Journal of Dynamic Systems, Measurement and Control. Vol. 122, No. 3, pp. 402-409, September 2000. [10] K. Krishnaswamy* and P.Y. Li, “Single Degree of Freedom Passive Hydraulic Teleoperation Using a Passified Two-Stage Servo-Valve,” 2nd IFAC Conference on Mechatronic Systems, Berkeley, CA, December 2002. [11] P.Y. Li and R.F. Ngwompo, “Power Scaling Bond Graph Approach to the Passification of Mechatronic Systems – With Application to Electrohydraulic Valves,” ASME Journal of Dynamic Systems, Measurement and Control. Vol. 127, No. 4, pp. 633-641, December 2005. [12] P.Y. Li and K. Krishnaswamy, “Passive Bilateral Teleoperation of a Hydraulic Actuator Using an Electrohydraulic Passive Valve,” International Journal of Fluid Power. Vol. 5(2), pp. 43-56, August 2004. [13] P.Y. Li* and K. Krishnaswamy, “Towards Human Friendly Hydraulics - Passive Teleoperation of Hydraulic Equipment Using a Force Feedback Joystick,” In Proceedings of the 49th National Conference on Fluid Power, NCFP-I02-26.2/SAE-OH-2002-01-1492, pp. 639-650, Las Vegas, NV, March 2002. Also in SAE 2002 Transactions - Journal of Commercial Vehicles, pp. 495-505, 2003. [14] K. Krishnaswamy and P.Y. Li, “Bond Graph Based Approach to Passive Teleoperation of a Hydraulic Backhoe,” ASME Journal of Dynamic Systems, Measurement and Control. Vol. 128, No. 1, pp. 176-185, March 2006. [15] P.Y. Li*, “Design and Control of a Hydraulic Human Power Amplifier,” IMECE 2004-60868, Anaheim, CA, November 2004. [16] P.Y. Li, “A New Passive Control for Hydraulic Human Power Amplifier,” ASME-IMECE200615056, Chicago, IL, November 2006. [17] K. Krishnaswamy and P.Y. Li, “On Using Unstable Electrohydraulic Valves for Control,” ASME Journal of Dynamic Systems, Measurement and Control. Vol. 124, No. 1, pp. 183-190, 2002. [18] Q.H. Yuan and P.Y. Li, “Using Steady Flow Force for Unstable Valve Design: Modeling and Experiments,” ASME Journal of Dynamic Systems, Measurement and Control. Vol. 127, No. 3, pp. 451-462, September, 2005. [19] Q.H. Yuan and P.Y. Li, “Robust Optimal Design of Unstable Valves,” IEEE Transactions on Control Systems Technology. Vol. 15, No. 5, pp. 1065-1074, November, 2007. [20] Q. H. Yuan, Unstable Electrohydraulic Valves, PhD Dissertation, Department of Mechanical Engineering, University of Minnesota, 2006. [21] C. Groepper, T. Cui, P.Y. Li*, K.A.Stelson, “Fabrication of Integrated Pressure-Flow-Temperature Sensor for Hydraulic Systems”, ASME-IMECE #13211, Chicago, November, 2006. [22] C. Groepper, T. Cui, P.Y. Li, K.A.Stelson*, “Design of Integrated Pressure Flow and Temperature Sensor for Hydraulic Systems”, In Proceedings of the Bath Workshop for Power Transmission and Motion Control (PTMC). N. Johnston and K. A. Edge, Eds., Professional Engineering Publishing Ltd. London. pp. 321-334, September, 2006. [23] P.Y. Li* and Q.H. Yuan, “Flux Observer for Spool Displacement Sensing in Self-Sensing Push-Pull Solenoids,” 5th International Conference on Fluid Power (ICFP), Hangzhou, China, April 2005. [24] Q.H. Yuan and P.Y. Li, “On Self-Sensing Actuators: Comparison of Boxcar Window and Kalman Filter Approaches,” ASME-IMECE, #IMECE2005-80284, Fluid Power Systems and Technology Division, Orlando, FL, November 2005. [25] M. Rannow, H. Tu, P.Y. Li, T.R. Chase and C.Y. Li, “Software Enabled Variable Displacement Pumps – Modeling and Experiments,” ASME Journal of Dynamic Systems, Measurement, and Control. (under review) [26] M. Rannow*, H. Tu, P.Y. Li and T.R. Chase, “Software Enabled Variable Displacement Pumps – Experimental Studies,” ASME IMECE2006-14973, Chicago, IL, November 2006.

8

Perry Y. Li [27] H. Tu, M. Rannow, M. Wang, J. Van de Ven, P.Y. Li and T.R. Chase, “High Speed Rotary Pulse Width Modulated On/Off Valve,” ASME-IMECE, Seattle, WA, November 2007. [28] P. Y. Li, “Open Accumulator Approach for Compact Energy Storage”, ASME-IMECE, Seattle, November, 2007. [29] J. D. Van de Ven, M. Olson and P.Y. Li, “Development of a Hydro-Mechanical Hydraulic Hybrid Drive Train with Independent Wheel Torque Control for an Urban Passenger Vehicle,” IFPE, Las Vegas, NV, March, 2008. [30] P.Y. Li*, T.-P. Sim and D.J. Lee, “Time Sequential Sampling and Reconstruction of Tone and Color Reproduction Functions for Xerographic Printing,” The 2004 American Control Conference, Boston, MA, Vol. 3, pp. 2630-635, June 2004. [31] T.P. Sim, P.Y. Li and Dongjun Lee “Using Time-Sequential Sampling to Stabilize the Color and Tone Reproduction Functions of a Xerographic Printing Process,” IEEE Transactions on Control Systems Technology. Vol. 15, No. 2, pp. 349-357, 2007. [32] T.P. Sim* and P.Y. Li, “Optimal Time Sequential Sampling of Tone/Color Reproduction Functions,” 2006 American Control Conference, Minneapolis, MN, pp. 5278-5233, June 2006. [33] T.-P. Sim* and P.Y. Li, “Color Reproduction Stabilization with Time-Sequential Sampling,” NonImpact Printing (NIP-20), Salt Lake City, UT, October 2004. [34] P.Y. Li, S. Ramaswamy and P. Bjegovic, “Preemptive Control of Moisture Content in Paper Manufacturing Using Surrogate Measurements,” Transactions of the Institute of Measurement and Control. Vol. 25, No. 1, pp. 36-56, 2003. [35] P. Bjegovic and P.Y. Li*, “Adaptive Pre-Emptive Control of Vacuum Dewatering in Paper Manufacturing,” 2002 IFAC World Congress, Barcelona, Spain, June 2002. [36] Z. Liu* and P.Y. Li, “Adaptive Calibration of Surrogate Measurements and its Application to Preemptive Control of Moisture Content in Paper Manufacturing,” Proceedings of the 2004 American Control Conference, Boston, MA, Vol. 4, pp. 3203-3208, June 2004.

9

Perry Y. Li

Research Grants and Gifts at the U. of Minnesota Source Federal / State National Science Foundation

National Science Foundation

National Science Foundation

National Science Foundation National Science Foundation National Science Foundation

National Science Foundation National Science Foundation National Science Foundation Minnesota Dept. of Transportation / Ctr. for Transportation Studies

Project Title

Award Amount

Engineering Research Center (ERC) for Compact and Efficient Fluid Power K. Stelson (Director), P. Y. Li (Deputy director)

approx 6/06-5/11 $21,000,000 (incl. $6M from industry and U’s match) 2/07-1/12

Industry/University Collaborative Research Center (I/UCRC): Minimally Invasive Medical Technologies Center, A. G. Erdman (PI), P. Y. Li (co-PI) Planning Grant: Industry/University Collaborative Research Center (I/UCRC) for Minimally Invasive Technologies, A. G. Erdman (PI), P. Y. Li (co-PI) Software enabled variable displacement pumps, P. Y. Li (PI), T. R. Chase (Co-PI) Distributed Laboratory for Systems Dynamics and Control, W. K. Durfee (PI), P. Y. Li (Co-PI) Collaborative Research: Sensing and Control for the Digital Xerographic Printing Process. P. Y. Li (PI) [Collaboration with Purdue] Instability and Passivity Concepts for Fast and Safe Electrohydraulic Systems Control of Moisture Content for Paper Manufacturing. P. Y. Li (PI), S. Ramaswamy (Co-PI) Passive Control of Mechanical Systems with Coordination Requirements Using Adaptive Cruise Control (ACC) vehicles to control traffic flow on semiautonomous highways

10

Period of Award

$300,000 8/05-8/06 $9,999 9/04-8/08 $340,000 4/03-3/05 $75,000.00 $188,357

8/02-7/07

$180,000

9/00-8/04

$325,000

9/00-8/04

$150,000

6/98-5/02

$89,386 (direct cost)

12/98-12/00

Perry Y. Li

Industry Cummins Power Generation National Fluid Power Association – Cooperative Network for Research (CNR) Xerox Education Foundation National Fluid Power Asscoiation National Fluid Power Asscoiation National Fluid Power Asscoiation IRIS Corporation

Adaptive Control of Gensets

$110000

3/06-9/07

Integrated P-Q-T Hydraulic Sensor K. A. Stelson (PI), P. Y. Li and T.H. Cui (co-PIs)

$100000

9/04-8/06

Statisical control and diagnostics of print engines Control Research Experience for Undergraduate Students R/C model engine powered hydraulic cylinders. W. Durfee (PI), P. Y. Li (co-PI) Hydraulics Exoskeleton

$60000 (gift)

6/04-5/07

$5000 (gift)

9/04-8/05

$5000 (gift)

9/04-8/05

$5000 (gift)

9/03-8/04

Development of a Benchtop Dental Model Scanner. A. Erdman (PI), P. Y. Li (Co-PI) TAPPI Foundation Process control for the pulp and paper industry: pre-emptive and adaptive strategy, and the use of surrogate measurements. P. Y. Li (PI), S. Ramaswamy (Co-PI) A Consortium of MN Fluid Power Education and Research Fluid Power Co. (Eaton, Initiatives at the U. of Minnesota. P. Y. Li Toro, Sauer-Danfoss, (Director), K. Stelson (Co-director) MTS) Xerox Education Diagnostics and Control of the Foundation Xerographic Printing Process Others University of Minnesota Design and Control of Human Powered (Grant-in-Aid) Swimming Machines

$88,127 (direct cost) $40,000 (direct cost)

1/98-8/00

Approx. $90,000 (gift) + in kind donations $35,000 (gift)

1/99-12/00

$20,120

12/98-12/99

University of Minnesota Machines that co-operate with Humans (Grant-in-Aid)

$15,120

12/97-12/98

Total funding at the University of Minnesota: 1,741,110

(as PI, excluding ERC) (as co-PI. excluding ERC) (total excluding ERC)

11

8/99-7/00

6/99-6/01

$ $ 490,000 $ 2,231,110

Perry Y. Li Grants at Xerox Corporation Source Project title National Science Foundation

GOALI: Mechatronic Design and Control of Media Handling Systems for Printing Engines. ENG/CMS-9632828 M. Tomizuka (PI), R. Horowitz (co-PI), P. Y. Li (industry co-PI)

12

Award Amount $197,426

Award Period 9/96-8/99