Optimization Online


Stability of Polynomial Differential Equations: Complexity and Converse Lyapunov Questions

Amir Ali Ahmadi(a_a_a***at***mit.edu)
Pablo A. Parrilo(parrilo***at***mit.edu)

Abstract: We consider polynomial differential equations and make a number of contributions to the questions of (i) complexity of deciding stability, (ii) existence of polynomial Lyapunov functions, and (iii) existence of sum of squares (sos) Lyapunov functions. (i) We show that deciding local or global asymptotic stability of cubic vector fields is strongly NP-hard. Simple variations of our proof are shown to imply strong NP-hardness of several other decision problems: testing local attractivity of an equilibrium point, stability of an equilibrium point in the sense of Lyapunov, invariance of the unit ball, boundedness of trajectories, convergence of all trajectories in a ball to a given equilibrium point, existence of a quadratic Lyapunov function, local collision avoidance, and existence of a stabilizing control law. (ii) We present a simple, explicit example of a globally asymptotically stable quadratic vector field on the plane which does not admit a polynomial Lyapunov function (joint work with M. Krstic). For the subclass of homogeneous vector fields, we conjecture that asymptotic stability implies existence of a polynomial Lyapunov function, but show that the minimum degree of such a Lyapunov function can be arbitrarily large even for vector fields in fixed dimension and degree. For the same class of vector fields, we further establish that there is no monotonicity in the degree of polynomial Lyapunov functions. (iii) We show via an explicit counterexample that if the degree of the polynomial Lyapunov function is fixed, then sos programming may fail to find a valid Lyapunov function even though one exists. On the other hand, if the degree is allowed to increase, we prove that existence of a polynomial Lyapunov function for a planar or a homogeneous vector field implies existence of a polynomial Lyapunov function that is sos and that the negative of its derivative is also sos.

Keywords: Stability of differential equations, sum of squares optimization, Lyapunov theory, semidefinite programming, NP-hardness

Category 1: Applications -- Science and Engineering (Control Applications )

Category 2: Linear, Cone and Semidefinite Programming (Semi-definite Programming )

Category 3: Convex and Nonsmooth Optimization (Convex Optimization )

Citation: 30 pages. Submitted for journal publication.

Download: [PDF]

Entry Submitted: 09/01/2013
Entry Accepted: 09/01/2013
Entry Last Modified: 09/01/2013

Modify/Update this entry

  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository


Coordinator's Board
Classification Scheme
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society