TY - CHAP
T1 - Control of Nonlinear Descriptor Systems A Computer Algebra Based Approach
AU - Schlacher, Kurt
AU - Kugi, Andreas
PY - 2000/6
Y1 - 2000/6
N2 - Many problems in mathematical modeling of lumped parameter systems lead to sets of mixed ordinary differential and algebraic equations. A natural generalization are so called descriptor systems. Based on the presented geometric framework using the mathematical language of Pfaffian systems, we derive a canonical form of a descriptor system under some mild rank conidtions. This form is equivalent to an explicit system, whenever some integrability conditions are met. This approach allows us to extend the well known concepts of accessibility, observability, equivalence by static feedback, etc., to the class of descriptor systems. The Euler-Lagrange and Hamilton-Jacobi equations for optimal control problems with descriptor systems are also derivable from this canonical form similar to the case of explicit control systems. In addition, this approach offers computer algebra based algorithms, which permit to apply the presented methods efficiently to real world problems.
AB - Many problems in mathematical modeling of lumped parameter systems lead to sets of mixed ordinary differential and algebraic equations. A natural generalization are so called descriptor systems. Based on the presented geometric framework using the mathematical language of Pfaffian systems, we derive a canonical form of a descriptor system under some mild rank conidtions. This form is equivalent to an explicit system, whenever some integrability conditions are met. This approach allows us to extend the well known concepts of accessibility, observability, equivalence by static feedback, etc., to the class of descriptor systems. The Euler-Lagrange and Hamilton-Jacobi equations for optimal control problems with descriptor systems are also derivable from this canonical form similar to the case of explicit control systems. In addition, this approach offers computer algebra based algorithms, which permit to apply the presented methods efficiently to real world problems.
UR - http://regpro.mechatronik.uni-linz.ac.at
M3 - Chapter
SN - 1-85233-364-2
VL - 259
T3 - Lecture Notes in Control and Information Sciences
SP - 379
EP - 395
BT - Nonlinear Control in the Year 2000
A2 - Isidori A., Lamnabhi-Lagarrigue F., Respondek W., null
PB - Springer Verlag
ER -