Year: 2026
Type: article
Abstract: In optimal control, extending the class of admissible controls is a common strategy to guarantee the existence of optimal solutions. However, such extensions may introduce a gap between the infimum of the original problem and the minimum of the extended one, especially in the presence of endpoint constraints. Since Warga’s seminal work – where the extension consisted of the convexification of the dynamics –, normality of first-order necessary conditions for extended minimizers has been recognized as a sufficient condition to avoid the gap phenomenon, though it is far from being necessary. In this paper, we consider impulsive extensions of control-affine systems with unbounded controls. We prove that a notion of higher-order normality , based on iterated Lie brackets of the system’s vector fields, is sufficient to prevent an infimum gap. Our main contribution is to establish this result for local minimizers with respect to the L 1 -distance between controls, rather than the L ∞ -topology on trajectories considered in previous higher-order no-gap results. This significantly strengthens the theory, since L 1 -control-local optimality is a weaker notion. However, this generalization does not apply to arbitrary extensions: indeed, Vinter exhibited a counterexample for relaxation-based extensions where first-order normality with respect to the L 1 -topology on controls does not exclude an infimum gap. Our method relies on a set-separation approach, which allows us to derive higher-order conditions and exploit the corresponding notion of higher-order normality.
Source: Systems & Control Letters
Institutions: University of Padua, University of L'Aquila
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,910
Award: G5245466801
OpenAlex ID: W7134227766
