Jie Sheng, PhD

Assitant Professor

Institute of Technology

University of Washington Tacoma

Contact: shengj2@u.washington.edu

Teaching:

Courses in Spring 2010:

TCSS142 – Introduction to Object-Oriented Programming

TCES482 – Senior Project II

Courses taught since Winter 2009:

TCES481 – Senior project I

TCES482 – Senior project II

TCES430 – Microprocessor System Design

TCES230 – Introduction to Logic Design

TCES390 – Undergraduate Seminar in CES

TCSS390 – Undergraduate Seminar in CSS

Research:

Research interests: Microprocessor embedded systems, robotics, signals and systems, engineering education

Selected journal publication:

· Y. Liu, J. Sheng, and R. Ding, Convergence of stochastic gradient estimation algorithm for multivariable ARX-like systems, to appear in Computers and Mathematics with Applications, 2010.

· L. Han, J. Sheng, F. Ding, and Y. Shi, Auxiliary model identification method for multirate multi-input systems based on least squares, Mathematical and Computer Modeling, vol. 50, pp. 1100-1106, 2009.

· Y. Zhang, Y. Chen, J. Sheng, and T. Hesketh, Fault detection and diagnosis of networked control system, International Journal of Systems Science, vol. 39, issue 10, pp. 1017-1024, 2008.

· J. Sheng, T. Chen and S. L. Shah, Optimal filtering for multirate systems, IEEE Transactions on Circuits and Systems, Part II: Analog and Digital Signal Processing, vol. 52, pp. 228-232, 2005.

· J. A. Rossiter, J. Sheng, T. Chen and S. L. Shah, Interpretations of and options in dual-rate predictive control, Journal of Process Control, vol. 15, pp. 135-148, 2005.

· J. Sheng, T. Chen and S. L. Shah, Generalized predictive control for non-uniformly sampled systems, Journal of Process Control, vol. 12, pp. 875-885, 2002.

Recent conference publication:

· J. Sheng, L.L. Wear, and O. R. Baiocchi, Computer Engineering and Systems Capstone Design Course at UW Tacoma, to appear in Proc. of National Capstone Conference 2010, June 7-9, Boulder, Colorado.

· H.K. Ra, S. Chung, and J. Sheng, Teaching the Way of Using Computers with Autonomous Robots for Junior-High Students, to appear in Proc. of the 7th International Conference on Cybernetics and Information Technologies, Systems and Applications (CITSA 2010), June 29th - July 2nd, Orlando, Florida, USA.

· H. Zhang, A. S. Mehr, Y. Shi, and J. Sheng, New Results on Robust L2 − L∞ Filtering for Uncertain Linear Discrete-Time Systems, to appear in Proc. 2010 American Control Conference, Baltimore, Maryland, USA, June 30 - July 2, 2010.

· L.L. Wear, O. R. Baiocchi, and J. Sheng, Project-Oriented Courses for Freshmen Engineers, Proc. of the First Ibero-American Symposium of Project Approaches in Engineering Education (PAEE´2009), University of Minho in Portugal, July 21-22, pp. 31-35, 2009.

· L. Han, J. Sheng, F. Ding, Y. Shi, Recursive least squares identification for multirate multi-input single-output systems, Proc. 2009 American Control Conference, St. Louis, Missouri, pp 5604 – 5609, June 10-12, 2009.

· J. Sheng, Infinite Breakpoints and Breaking Using an External Trigger on the PIC18F4550, Proc. 2009 International Conference on Embedded Systems and Applications (ESA'09), Las Vegas, July 13-16, 2009.

Current research projects:

1. Using Open Source Software to Develop a Testbed for Unmanned Vehicle Systems with Smartphone as Communication Media

Unmanned Ground Vehicles (UVSs) are uninhabited, unpiloted, and remotely controlled vehicles that can be equipped to perform a variety of tasks including security, hazardous waste cleanup, monitoring of agricultural crops, law enforcement, as well as military operations. Travelling on land, in the sea or in the air, UVSs come in a wide variety of configurations. Some are small and tactical, while others are large and strategic. This diversity has naturally led to the interoperability problem between UVSs developed by difference organizations or companies. Aiming at an open architecture for UVSs, Joint Architecture for Unmanned Systems (JAUS) was originally initiated by the Department of Defense and is now maintained by the unmanned technical committee of the Society of Automotive Engineers . An emerging standard for messaging within and between UVSs, JAUS enhances code modularity and enables code reuse, leading to more rapid, less costly development and decreased deployment expense.

For autonomous UVSs, communication is necessary and essential in accomplishing complex mission tasks; it becomes especially important in cooperative missions where tasks are solved using many vehicles of different sizes and characteristics with different sensor suites. Smartphone technology offers an interesting communication infrastructure for remotely accessing, controlling and interacting with UVSs in an integrated and highly portable manner, and offers the ability to have an interface to the World Wide Web (WWW) for additional information useful in mission achievement. Smartphone technology is mature enough to begin experimentation and research into their integration with UVSs and their use in complex mission scenarios.

The purpose of this research is to apply open source software to the development of a testbed for UVSs with smartphone as communication media; within and between UVSs, JAUS standard would be implemented for message passing. The feasibility of using open source software, namely, OpenJAUS , in designing, developing and deploying UVSs will be investigated; Google Android will be used to build applications such as streaming video from UVSs, sending snapshots of targets to the central station, via smartphone; the performance robustness and security issues will also be examined.

2. Hardware Implementation of Multi-rate Model Predictive Control

Nowadays, micro-needles have been created to painlessly cross the uppermost layer of the skin to effectively, efficiently, and painlessly deliver drugs to depths as shallow as 1mm. It is the trend to develop small and compact drug delivery devices that can maximize the therapeutic results of a drug. Considering the human organism, which is one of the most complex dynamical systems, the drug delivery system needs an advanced control scheme to handle the inherent existence of nonlinearities, constraints, patient variability, multi-rates, and delays. Multi-rate MPC is a suitable candidate.

One example where multi-rate concept is useful is the drug delivery control problem to regulate the mean arterial pressure (MAP) and cardiac output (CO) for heart attack patients. CO measurements typically are available every 2-3 minutes, and MAP can be measured as often as every 30 seconds. The need for an explicit multi-rate control scheme arises because the settling time criteria may not be met if all measurements are slowed down to correspond to the CO sampling rate. Model Predictive Control (MPC) was originally developed in the late 1970s. More recently, there have been considerable interests in expanding the applicability of MPC to systems which were traditionally considered unsuitable for MPC due to their small physical size and fast dynamics. As an example, MPC can be used to regulate the blood glucose concentration of a diabetic by injecting insulin according to dynamic measurements of glucose concentration.

Advances in the semiconductor technology and in computer architecture have enabled the integration of enough computational power on a chip, to solve efficiently MPC problems in real time. Motivated by all these research achievement, and the need for a scalable and low-cost embedded multi-rate MPC algorithm for applications like the drug delivery system, this research will focus on the hardware implementation of multi-rate MPC. The research mainly has two objectives. The first objective is to implement the multi-rate MPC on general-purpose microprocessors. The second objective is to implement multi-rate MPC onto FPGAs (Field-programmable gate array).