on November 27, 2025 at 10:30 am
Small-Gain Conditions for Exponential Incremental Stability in Feedback Interconnections
Mohamed Yassine Arkhis (Inria)
Abstract: We prove that under a small-gain condition, an affine interconnection of two globally incrementally exponentially stable systems inherits this property on any compact connected forward invariant set. It is also demonstrated that the interconnection inherits a weaker version of incremental exponential stability globally. An example illustrating the theoretical findings is given. The example also shows that the uniform negativity of the Jacobian is not necessary for incremental exponential stability.
Dynamics of theta-invariant systems and normed actions
Danilo Lima (Inria)
Abstract: Symmetry methods have been largely used in the analysis of differential equations. A relevant class within symmetric systems is that of homogeneous systems, both for their global properties and their fast asymptotic convergence. In a previous work, we introduced the class of theta-invariant systems as a way to generalize the well-known concept of G-invariance to a larger symmetric class that allows for homogeneity of arbitrary degree. This talk is intended to present dynamical properties expected from invariant and theta-invariant systems and introduce the concept of an invariant norm, which generalizes homogeneous-like properties to this larger class.
Robust Interval Predictor-based Controller Design for Anesthesia Machines Using Two Drugs
Manuel Mera (IPN, Mexico)
Abstract: We present a novel interval-predictor-based controller to solve the regulation problem for anesthesia machines used in anesthetic procedures, where Propofol and Remifentanil are administered to the patient, and the only indicator of the depth of Hypnosis (DoH) is given by the Bispectral Index (BIS). The resulting control algorithm does not require the design of an observer or a filter to estimate the drug concentration levels on the patient’s effect site from the BIS signal; it only requires knowing an interval of possible initial conditions to ensure the convergence of the regulation errors to some target compact sets around the origin, despite perturbations affecting the patient’s pharmacodynamics and pharmacokinetics. Some simulations, considering different in silico patients, are presented to illustrate the implementability of the controller.
Inria salle Plénière
français
English