Searcharxiv⌕ Search

arXiv subjects

M. Soledad Aronna

Publications and source records attributed to M. Soledad Aronna.

21 records · Page 2Linked to original sources

A note on systems with ordinary and impulsive controls

We investigate an everywhere defined notion of solution for control systems whose dynamics depend nonlinearly on the control $u$ and state $x,$ and are affine in the time derivative $\dot u.$ For this reason, the input $u,$ which is allowed to be Lebesgue integrable, is called impulsive, while a second, bounded measurable control $v$ is denominated ordinary. The proposed notion of solution is derived from a topological (non-metric) characterization of a former concept of solution which was given in the case when the drift is $v$-independent. Existence, uniqueness and representation of the solution are studied, and a close analysis of effects of (possibly infinitely many) discontinuities on a null set is performed as well.

math.OC↗

On optimal control problems with impulsive commutative dynamics

We consider control systems governed by nonlinear O.D.E.'s that are affine in the time-derivative du/dt of the control u. The latter is allowed to be an integrable, possibly of unbounded variation function, which gives the system an impulsive character. As is well-known, the corresponding Cauchy problem cannot be interpreted in terms of Schwartz distributions, even in the commutative case. A robust notion of solution already proposed in the literature is here adopted and slightly generalized to the case where an ordinary, bounded, control is present in the dynamics as well. For a problem in the Mayer form we then investigate the question whether this notion of solution provides a "proper extension" of the standard problem with absolutely continuous controls u. Furthermore, we show that this impulsive problem is a variational limit of problems corresponding to controls u with bounded variation.

math.OC↗

Singular solutions in optimal control: second order conditions and a shooting algorithm

In this article we study optimal control problems for systems that are affine in one part of the control variable. Finitely many equality and inequality constraints on the initial and final values of the state are considered. We investigate singular optimal solutions for this class of problems. First, we obtain second order necessary and sufficient conditions for weak optimality. Afterwards, we propose a shooting algorithm and show that the sufficient condition above-mentioned is also sufficient for the local quadratic convergence of the algorithm.

math.OC↗