Set-valued calculus and dynamic programming in problems of feedback control

Publication date 28 November 1999

dynamic programming ellipsoidal calculus feedback control strategy linear control systems Pontryagin's alternating integral uncertainty

Mathematics Subject Classification ID

Set-valued and variational analysis (49J53) Dynamic programming in optimal control and differential games (49L20) Viscosity solutions to Hamilton-Jacobi equations in optimal control and differential games (49L25) Optimal feedback synthesis (49N35)

Review text

This survey contains the statements of a large number of results (some 10 lemmas and 4 theorems) concerning characterizations of ``feedback control strategies \({\mathcal U}(.,.)\) and ``solvability tubes \(W(.)\) such that all the solutions \(x(.)\) of the differential inclusion \[ x'\in A(t)x+{\mathcal U}(t,x)+v(t), \;t\in [\tau,t_1], \;x(\tau)\in W(\tau), \;\tau \in [t_0,t_1) \] would reach at \(t_1\) a given ``terminal set \({\mathcal M}\subset R^n\) (i.e., \(x(t_1)\in {\mathcal M}\)) whatever the ``disturbances \(v(t)\in Q(t)\). Associating certain ``cost functionals, the corresponding value functions (which turn out to be viscosity solutions of the Bellman-Isaacs equation) and using the so-called ``ellipsoidal calculus the author provides some characterizations of an ``internal ellipsoidal approximation \({\mathcal E}_-(.)\) of the solvability tube \(W(.)\) and of a corresponding ``synthesizing strategy \({\mathcal U}_-(.,.)\) that solve the problem for points \(x(\tau)\in {\mathcal E}_-(\tau) \subset W(\tau)\), \(\tau \in [t_0,t_1)\). For the entire collection see [Zbl 0885.00052].

Recommendations

Cited in

(5)

This page was built for publication: Set-valued calculus and dynamic programming in problems of feedback control

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4249811)