Xiang Cheng

Associate Professor

Ph.D. Yale University; 2000
M.S. Inst. Mechanics, Chinese Academy of Sciences; 1994
B.S. Fudan University; 1991

 (65)65166210 Click here to Email

Professional Working Experience

		
Financial Engineer, Fannie Mae, Washington D.C., U.S.A., 2000-2001.

Research Assistant, Center for Systems Science, Yale University, 1995-2000
		
		

Professional Activities

Serve as the chairman of the IEEE Singapore Control Systems Chapter since 2010.

Serve as the Vice chairman of the IEEE Singapore Control Systems Chapter, 2008-2009.

Serve as the chairman for the 1st, 2nd and 3rd IEEE Singapore Control Systems Chapter 
Graduate Student Workshop in Control and Automation, in 2005, 2006 and 2007.

Serve as the Finance Chair of the inaugural 2007 IEEE Multi-conference on Systems and 
Control (IEEE MSC2007). 

Serve as the Committee member of the IEEE Singapore Control Systems Chapter, 2006-2007.

Serve as the Secretary of the IEEE Singapore Control Systems Chapter 2003-2005.

Serve in the organizing committees for ICCA05, ICCA07, ICCA09, ICCA10 and ICCA11.

Serve in the organizing committees for CIRAS 2003 and CIRAS 2005.

Serve in the international program committees for ISNN2006, ISNN2007, ISNN2008, 
LSMS2007, and CIS-RAM 2008.

	
		
		

Research Interest

The mind, not space, is science's  final frontier, which stimulates both scientists and 
engineers to discover how the human brain works and how to make machines really 
intelligent just like human beings that can learn from experience and improve their 
performances. The long term goal of my research programme is to investigate the 
fundamental issues related to the three most important aspects of the theme of 
intelligent systems and machines: pattern recognition, control and learning.  Rather 
than limiting my interest to only one aspect, I am keeping my mind open to all possible 
interesting fundamental issues relevant to these three essential features. 

Although significant progress has been made in understanding the various fundamental 
issues in building intelligent systems and machines, there are plenty of open questions 
for us to investigate in the future. For instance, we have just shifted the paradigm for 
pattern recognition from the statistical approach to biological approach since the 
statistical approach has proven to be almost hopeless for pattern recognition problems 
where the training examples are scarce or the within-class variance is large while even 
the toddlers or the monkeys can learn to recognize different objects from very few 
examples with large variances. We are currently taking on the challenging problem of how 
to incorporate the various fundamental features of the human vision system together into 
one vision machine synergistically. 

From the very beginning, the primary motivation for utilizing multiple models, switching 
and tuning for adaptive control is to deal with time-varying environment.  However, due 
to mathematical tractability, stability analysis has been confined to time-invariant 
systems with unknown parameters while very few analytical results have been reported on 
how to identify and control the system in rapidly changing environment except some 
heuristic ideas. The first question we would like to address along this line is how to 
locate the models in a rapidly time-varying environment. Most recently, we have 
discovered a general framework for identification of discrete-time time-varying systems. 
Extensive simulation studies have shown that our algorithm can indeed provide accurate 
estimates of the plant parameters even in noisy cases.  Once the identification problem 
is solved satisfactorily, we will be ready to attack the challenging problem of adaptive 
control of time-varying systems using multiple models, which will surely make major 
impact once accomplished.

We have just embarked on the adventure in exploring the feasibility of applying feedback 
control to regulate gene activities. The central tool for the gene engineering community 
is deleting or inserting the genes in existing networks. In contrast to this main 
stream, we are taking a different approach by investigating the feasibility and 
efficiency in utilizing feedback mechanisms to turn genes on and off without changing 
their presence. The motivation of this approach is based upon the biological fact that 
the highly specialization of different cells can be traced to the different activation 
levels of the various genes inside the cells despite the fact that all the cells share 
the same genes within the same living beings. Currently, we are trying hard to conduct 
the experimental work to validate this approach. 		
		
		

Research Projects

Adaptive Control using Multiple Models  (PI)     
NUS ARF, S$37,290,  01/04/2002 to 31/03/2006    ---- --- completed 

Pattern Recognition: A Biological Approach (PI)   
NUS ARF, S$146,250,  01/02/2006 to 28/02/2009---- --- completed

Experimentation, Modeling and Control of Calcium Dynamics in Human Vascular Endothelial 
Cells (PI)		
NUS ARF, S$154,800,  01/02/2008 to 31/03/2011 ---- --- completed


Coordination and Control of Multi-Robot Systems: Hybrid System Approaches (PI)
Temasek Labs, S$200,000,  01/08/2009 to  31/07/2013 ongoing



STARFISH (Small Teams of Autonomous Robtic Fish) Robust Positioning and Localization (Co-
PI)
NUS ARF, S$122,000,  01/04/2006 to 30/09/2008 ---- --- completed

Social Robots: Breathing Life into Machines  (Collaborator)
MDA, S$1,597,800,  2007 to 2011 ---- --- completed


		
		
		

Selected Publications

1. W. Gu, C. Xiang, Y. V. Venkatesh, D. Huang and H. Lin, ˇ°Facial Expression 
Recognition using Radial Encoded Local Gabor Features and Classifier Synthesis,ˇ± 
Pattern Recognition, vol. 45, no. 1, pp. 80-91, 2012.
2. C.Y. Lai, C. Xiang and T.H. Lee, "Data-based Identification and Control of Nonlinear 
Systems via Piecewise Affine Approximation,ˇ± IEEE Trans.  on Neural Networks, vol. 22, 
no. 12, pp.2189¨C2200, 2011.
3. K.R. Qin and C. Xiang, "Hysteresis Modeling for Calcium mediated Ciliary Beat 
Frequency in Airway Epithelial Cells," Mathematical Biosciences,  vol. 229, no. 1 , 
pp. 101-108, 2011.
4. Z. Huang, C. Xiang, H. Lin and TH Lee, "Necessary and Sufficient Conditions 
for Regional Stabilizability of Generic Switched Linear Systems with a Pair of Planar 
Subsystems," Int. J.  Control, vol. 83, no. 4 pp. 694-715, 2010.
5. KR Qin, C Xiang, Z. Xu, LL Cao, S S Ge and ZL Jiang, "Dynamic Modeling for 
Shear Stress Induced ATP Release from Vascular Endothelial Cells," Biomechanics and 
Modeling in Mechanobiology, vol. 7, no.5, pp. 345-353, 2008.
6. E. J. Teoh, KC Tan and C Xiang, "Estimating the number of hidden neurons in a 
feedforward network using the Singular Value Decomposition,"  IEEE Transactions on 
Neural Networks, vol. 17, no. 6, pp.1623-1629, 2006
7. C. Xiang and D. Huang, "Feature Extraction Using Recursive Cluster-Based 
Linear Discriminant with Application to Face Recognition,"  IEEE Trans. on Image 
Processing,vol. 15, no.12, pp.3824-3832, 2006. 
8. C. Xiang, X.A. Fan and T.H. Lee,"Face Recognition Using Recursive Fisher 
Linear Discriminant," IEEE Trans. on Image Processing, vol. 15, no. 8, pp. 2097-2105,   
2006. 
9. C. Xiang, S. Q. Ding and T.H. Lee, "Geometrical Interpretation and 
Architecture Selection of MLP,"  IEEE Transactions on Neural Networks, vol. 16, no. 1, 
pp. 84-96,  2005.
10. KS Narendra and C Xiang, "Adaptive Control of Discrete-time Systems Using 
Multiple Models," IEEE Transactions on Automatic Control, AC-45, no. 9, pp. 1669-1686, 
2000.
		
		
		

Awards

		
Yale University Fellowship, 1994-1995. Granted for outstanding academic achievement and 
performance.

Invited Speaker at the Fourteenth Yale Workshop on Adaptive and Learning Systems, Yale 
University, June 2-4, 2008. The Yale Workshop on Adaptive and Learning Systems is a 
prestigious workshop, whose speakers are by invitation only from the organizer.

Placed on Honors List for teaching, Faculty of Engineering,  NUS, 2009.
Placed on Commendation List for teaching, Faculty of Engineering, NUS, 2011.
	
		

Teaching Courses

EE3304 Digital Control Systems, S2
EE5103/ME5403 Computer Control Systems, S1
EE5904/ME5404 Neural Networks, S2