NOTE: in case Javascript or Style (CCS) has been turned off in your browser, the menus on this page will not work properly.

Research

The research interests of Raymond de Callafon include topics in the field of experiment-based approximation modeling, control relevant system identification and recursive/adaptive control. In particular, he is interested in designing and analyzing experiment-based modeling techniques for control relevant identification and model-based adaptive control and extending these techniques to specific classes of non-linear and linear parameter varying (LPV) systems. Applications of his research vary from complex engineering systems such as electromechanical (mechatronic) systems, structural engineering, aerospace and industrial processes in which signal processing, data based modeling, fault detection and feedback/feedforward control plays an important role.

The research is directed by Raymond de Callafon within the System Identification and Control Laboratory (SICL) at the Dept. of Mechanical and Aerospace Engineering at UCSD. A listing of current resarch projects is listed below. For more details on the research group of Raymond de Callafon please follow this page down to the description of the SICL.

Projects

Motion Control for Data Storage Systems

Close-up image of a piezo-electric dual-stage servo actuator in a Hard Disk Drive

In a continuing effort funded by the Information Storage Industry Consortium (INSIC) and a collaboration with the Center for Magnetic Recording Research (CMRR), Raymond de Callafon has been working on the data-based dynamic modeling and design of high performance mechatronic motion control systems used in high precision data storage devices. Applications include servo systems found in high-density magnetic Linear Tape Open (LTO) drives and hard disk drive (HDD) systems. The research is motivated by the need to increase storage density of magentic media, requiring dynamic models of dedicated (dual-stage) servo actuators and the design of a well-tuned (robust) servo control system. As part of this research effort, frequency-domain estimation technqiues for control relevant model estimation have been developed and bundled in a Matlab toolbox called LTSfit.

Adaptive Noise and Vibration Control

An active noise control headset being tested in a noise controlled environment

Closely related to the modeling and control of mechatronic systems is the reduction of vibrations and sound in infinite dimensional systems. Due to due to the lack of limited complexity physical models that can accurately describe sound and vibration propagation, data based modeling approaches are being developed in the System Identification and Control Laboratory (SICL). Both adaptive feedforward and feedback control algorithms haven been developed with application to sound cancellation in forced air cooling systems and vibration control in aeroelastic systems. Most recently, this research effort has also led to a Robust Estimation for Automatic Control Tuning (REACT) and publised in a monograph entitled "Realtime Controller Tuning for Periodic Disturbance Rejection", co-authored with Dr. Charles Kinney. Control of flutter induced vibrations in aero(servo)elastic systems has been funded by two subsequent NASA Graduate Student Research Fellowships and a recently awarded NASA Phase II proposal in cooperation with ZONA Technologies Inc. and NASA Dryden Flight Research Center.

Mechatronics and Coordinated Robotics

Development of a single wheel unibot robot that requires 2D stabilization

Initiated by a generous gift of Wowee Inc. and a recently funded Chancellor's Collaboratories Proposal on autonomous networked ocean sensing, the expertise of the System Identification and Control Laboratory (SICL) in the field of mechatronics and system identification for dynamic modeling has allowed the development of new robotic systems that require embedded estimation and control algorithms for modeling, stabilization, motion planning and coordination. Prototype robotic systems are created with the manufacturing capabilities of a 3-axis CNC machine at the SICL and a 2-axis LaserCamm at the Dept. of MAE. Most recently, an NSF CDI (Cyber-Enabled Discovery and Innovation) award (co-PI) in collaboration with the San Diego Supercomputer Center (SDSC) and the Scripps Institute for Oceanographic Research (SIO) is used to develop an integrated solution for oceanographic data analysis using a coordinated network of buoyancy controlled submersible drogues.

System Identification and Control Laboratory

Walk-in view of the SICL, located at room 303, EBU2, Dept. of MAE, UCSD

Raymond de Callafon direct the System Identification and Control Laboratory (SICL), located at room 303, Engineering Building II at the Dept. of Mechanical and Aerospace Engineering at UCSD. The SICL is a 800 square feet laboratory space that houses several facilities to conduct experiment and research related to sound and vibration control and estimation and control for mechatronic systems and coordinated robotics. Among the several experimental facilities, the SICL houses:

• an anechoic chamber for modeling and active control of sound,
• an active silencer for testing of adaptive sound cancellation techniques,
• a shaker table for vadaptve vibration control techniques,
• a 3-axis CNC machine for prototype manufacturing,
• computer hardware and software for embedded mircoprocessor (PIC and TI DSP) programming, and
• at least 5 computers with dedicated real-time hardware for real-time prototyping of control systems.

Several Ph.D, M.Sc. and (international) exchange students are affiliated with the SICL and (being) supervised by Raymond de Callafon. For more details on the research group of Raymond de Callafon or the experimental facilities of the SICL, please visit the webpage of the System Identification and Control Laboratory.

CMRR Servo Laboratory

Raymond de Callafon also manages the CMRR Servo Laboratory located at room 201, Center for Magnetic Recording Research (CMRR) at UCSD. The CMRR Servo Laboratory provides the experimental facilities to conduct research and experiments in advanced motion planning, servo control and experimental modeling of high precision servo systems in data storage applications. Current demands on Extremely High Density Recording (EHDR) with data storage capacities approaching 10 TerraBit per square inch for hard disk drives and 8 TerraByte/cartridge for tape storage systems have unleashed the need to develop dedicated servo control systems. The CMRR Servo Laboratory works closely with the Tribology and Mechanics Laboratory of Prof. Talke at the CMRR and houses:

• multiple Hard Disk Drive (HDD) experiments with special purposes embedded DSP boards developed by Headway for in-situ modeling, and control design
• an air-bearing spinstand for testing HDD dual-stage actuator modeling and control,
  
• multiple LTO tape drives and special purpose DSP hardware for in-drive measurements for actuator modeling and adaptive disturbance cancellation,
• single and two-axis LDV systems for high precision velocity and displacement measurements.

For more information on the research and experimental facilities, please visit the webpage of the Center for Magnetic Recording Research.