Galerkin approximation of the generalized Hamilton-Jacobi-Bellman equation.pdfVIP

GALERKIN APPROXIMATION OF THE GENERALIZED HAMILTONJACOBI EQUATION Randy Beard  1 George Saridis  John Wen   Electrical Computer and Systems Engineering Department Rensselaer Polytechnic Institute Abstract If u is a stabilizing control for a nonlinear system that is ane in the con trol variable then the solution to the Generalized HamiltonJacobiBellman GHJB equation asso ciated with u is a Lyapunov function for the system and equals the cost asso ciated with u If an explicit solution to the GHJB equation can b e found then it can b e used to construct a feedback control law that improves the p erformance of u Rep eating this pro cess leads to a successive approximation algorithm that uniformly approximates the HamiltonJacobiBellman equation The diculty is that it is very dicult to construct solutions to the GHJB equation such that the control derived from its solution is in feedback form This pap er shows that Galerkins approxima tion metho d can b e used to construct arbitrarily close approximations to the GHJB equation while generating stable feedback control laws We state sucient conditions for the convergence of Galerkin approximations to the GHJB equation The sucient conditions derived in this pap er include standard completeness assumptions and the asymptotic stability of the asso ciated vector eld The metho d is demonstrated on a simple nonlinear system and is compared to a result obtained by using exact feedback linearization in conjunction with the LQR design metho d