arXiv subjects

Fernando Castaños

Publications and source records attributed to Fernando Castaños.

12 recordsLinked to original sources

arXiv2025

Formalizing Neuromorphic Control Systems: A General Proposal and A Rhythmic Case Study

Neuromorphic control is receiving growing attention due to the multifaceted advantages it brings over more classical control approaches, including: sparse and on-demand sensing, information transmission, and actuation; energy-efficient designs and realizations in neuromorphic hardware; event-based signal processing and control signal computation. However, a general control-theoretical formalization of what "neuromorphic control systems" are and how we can rigorously analyze, design, and control them is still largely missing. In this note, we suggest a possible path toward formalizing neuromorphic control systems. We apply the proposed framework to a rhythmic control case study and rigorously show how it has the potential to make neuromorphic control systems analysis and design amenable to mature control theoretical approaches like describing function analysis and harmonic balance, fast-slow analysis, discrete and hybrid systems, and robust optimization.

eess.SY↗

arXiv2025

Sliding motions on systems with non-Euclidean state spaces: A differential-geometric perspective

This paper extends sliding-mode control theory to nonlinear systems evolving on smooth manifolds. Building on differential geometric methods, we reformulate Filippov's notion of solutions, characterize well-defined vector fields on quotient spaces, and provide a consistent geometric definition of higher-order sliding modes. We generalize the regular form to non-Euclidean settings and design explicit first- and second-order sliding-mode controllers that respect the manifold structure. Particular attention is given to the role of topological obstructions, which are illustrated through examples on the cylinder, M\"obius bundle, and 2-sphere. Our results highlight how geometric and topological properties fundamentally influence sliding dynamics and suggest new directions for robust control in nonlinear spaces.

math.OC↗

arXiv2023

Dominant-Pole Placement for Predictor Synthesis

This article analyzes the high-gain prediction approach for nonlinear input-delay systems. The problem is discussed in the light of weighted homogeneity and input-to-state stability. The canonical form for uniformly observable nonlinear systems allows tuning the spectrum of the linear part by multiplicity-induced dominance and ensures closed-loop system input-to-state stability using the descriptor method for Lyapunov-Krasovskii functionals. Due to the trade-off between delay and gain margin, a limitation of high-gain results for time-delay systems. The limitation is overcome by using a cascade of sub-predictors. A comparative analysis is also presented, showing that our proposal achieves a better trade-off between delay and gain margin.

math.OC↗

arXiv2023

Equivalence of Linear Complementarity Problems: Theory and Application to Nonsmooth Bifurcations

Linear complementarity problems provide a powerful framework to model nonsmooth phenomena in a variety of real-world applications. In dynamical control systems, they appear coupled to a linear input-output system in the form of linear complementarity systems. Mimicking the program that led to the foundation of bifurcation theory in smooth maps, we introduce a novel notion of equivalence between linear complementarity problems that sets the basis for a theory of bifurcations in a large class of nonsmooth maps, including, but not restricted to, steadystate bifurcations in linear complementarity systems. Our definition exploits the rich geometry of linear complementarity problems and leads to constructive algebraic conditions for identifying and classifying the nonsmooth singularities associated with nonsmooth bifurcations. We thoroughly illustrate our theory on an extended applied example, the design of bistability in an electrical network, and a more theoretical one, the identification and classification of all possible equivalence classes in two-dimensional linear complementarity problems.

eess.SY↗

arXiv2020

Observer-based predictor for a SIR model with delays

We propose an observer for a SIR epidemic model. The observer is then uplifted into a predictor to compensate for time delays in the input and the output. Tuning criteria are given for tuning gains of the predictor, while the estimation-error stability is ensured using Lyapunov-Krasovskii functionals. The predictor's performance is evaluated in combination with a time-optimal control. We show that the predictor nearly recovers the performance level of the delay-free system.

eess.SY↗

arXiv2019

A notion of equivalence for linear complementarity problems with application to the design of non-smooth bifurcations

Many systems of interest to control engineering can be modeled by linear complementarity problems. We introduce a new notion of equivalence between linear complementarity problems that sets the basis to translate the powerful tools of smooth bifurcation theory to this class of models. Leveraging this notion of equivalence, we introduce new tools to analyze, classify, and design non-smooth bifurcations in linear complementarity problems and their interconnection.

math.DS↗

arXiv2016

Implementing robust neuromodulation in neuromorphic circuits

We introduce a methodology to implement the physiological transition {between distinct neuronal spiking modes} in electronic circuits composed of resistors, capacitors and transistors. The result is a simple neuromorphic device organized by the same geometry {and exhibiting the same input--output properties as} high-dimensional electrophysiological neuron models. {Preliminary} experimental results highlight the robustness of the approach in real-world applications.

math.OC↗

arXiv2015

Passivity-based PI control of first-order systems with I/O communication delays: A complete sigma-stability analysis

The PI control of first-order linear passive systems through a delayed communication channel is revisited in light of the relative stability concept called sigma-stability. Treating the delayed communication channel as a transport PDE, the passivity of the overall control-loop is guaranteed, resulting in a closed-loop system of neutral nature. Spectral methods are then applied to the system to obtain a complete stability map. In particular, we perform the D-subdivision method to declare the exact sigma-stability regions in the space of PI parameters. This framework is then utilized to analytically determine the maximum achievable exponential decay rate of the system while achieving the PI tuning as an explicit function of the decay rate and the system parameters.

math.OC↗

arXiv2015

Discrete-Time Models for Implicit Port-Hamiltonian Systems

Implicit representations of finite-dimensional port-Hamiltonian systems are studied from the perspective of their use in numerical simulation and control design. Implicit representations arise when a system is modeled in Cartesian coordinates and when the system constraints are applied in the form of additional algebraic equations (the system model is in a DAE form). Such representations lend themselves better to sample-data approximations. An implicit representation of a port-Hamiltonian system is given and it is shown how to construct a sampled-data model that preserves the port-Hamiltonian structure under sample and hold.

eess.SY↗

arXiv2014

Min-max piecewise constant optimal control for multi-model linear systems

The present work addresses a finite-horizon linear-quadratic optimal control problem for uncertain systems driven by piecewise constant controls. The precise values of the system parameters are unknown, but assumed to belong to a finite set (i.e., there exist only finitely many possible models for the plant). Uncertainty is dealt with using a min-max approach (i.e., we seek the best control for the worst possible plant). The optimal control is derived using a multi-model version of Lagrange's multipliers method, which specifies the control in terms of a discrete-time Riccati equation and an optimization problem over a simplex. A numerical algorithm for computing the optimal control is proposed and tested by simulation.

eess.SY↗

arXiv2014

Robust Output Regulation of Linear Passive Systems with Multivalued Upper Semicontinuous Controls

The use of multivalued controls derived from a special maximal monotone operator are studied in this note. Starting with a strictly passive linear system (with possible parametric uncertainty and external disturbances) a multivalued control law is derived, ensuring regulation of the output to a desired value. The methodology used falls in a passivity-based control context, where we study how the multivalued control affects the dissipation equation of the closed-loop system, from which we derive its robustness properties. Finally, some numerical examples together with implementation issues are presented to support the main result.

eess.SY↗

arXiv2013

Pole-placement in higher-order sliding-mode control

We show that the well-known formula by Ackermann and Utkin can be generalized to the case of higher-order sliding modes. By interpreting the eigenvalue assignment of the sliding dynamics as a zero-placement problem, the generalization becomes straightforward and the proof is greatly simplified. The generalized formula retains the simplicity of the original one while allowing to construct the sliding variable of a single-input linear time-invariant system in such a way that it has desired relative degree and desired sliding-mode dynamics. The formula can be used as part of a higher-order sliding-mode control design methodology, achieving high accuracy and robustness at the same time.

math.OC↗