Control Of Dynamic Systems Research Articles

Constraint-following control for dynamic systems with comprehensive constraints: the generalized Udwadia–Kalaba approach

Abstract A novel Generalized Udwadia-Kalaba (GUK) control method is proposed in this paper to effectively deal with the constraint-following problem with comprehensive constraints, be it equality constraints or inequality constraints. Fully considering the complex application scenario in practice which encompassed both clear and feasible trajectory, the equality constraints and inequality constraints are simultaneously applied to design rational and reliable controller. Specifically, a state transformation function is introduced to describe the inequality constraints. Further, the dynamic model with comprehensive constraints is established based on the GUK theory. Then, a corresponding constraint-following control is presented to render the system performance, uniform boundedness and uniform ultimate boundedness. The effectiveness of the GUK control is verfied by the stability analysis based on Lyapunov minimax approach and experimental verification of unmanned underwater vehicle (UUV). The desired control performances of UUV, constraint-following and collision avoidance, can be achieved. As an illustrative application, the UUV is designed to navigate through narrow sea trenches. Meanwhile, through comparative analysis with the linear quadratic regulator (LQR) method and Udwadia-Kalaba (UK) method, the superiority of this approach is demonstrated.

Finite Elements with Switch Detection for numerical optimal control of nonsmooth dynamical systems with set-valued heaviside step functions

This paper develops high-accuracy methods for the numerical solution of optimal control problems subject to nonsmooth differential equations with set-valued Heaviside step functions. An important subclass of these systems are Filippov systems. By writing the Heaviside step function as the solution map of a linear program and using its optimality conditions, the initial nonsmooth system is rewritten into an equivalent Dynamic Complementarity System (DCS). The Finite Elements with Switch Detection (FESD) method (Nurkanović et al., 2024) was originally developed for Filippov systems transformed via Stewart’s reformulation into DCS (Stewart, 1990). This paper extends it to the above mentioned class of nonsmooth systems. The key ideas are to start with a standard Runge–Kutta method for the DCS and to let the integration step sizes to be degrees of freedom. Then, additional conditions are introduced to allow implicit but accurate switch detection and to remove possible spurious degrees of freedom when no switches occur. The theoretical properties of the FESD method are studied. The motivation for these developments is to obtain a computationally tractable formulation of nonsmooth optimal control problems. Numerical simulations and optimal control examples are used to illustrate the favorable properties of the proposed approach. All methods introduced in this paper are implemented in the open-source software package nosnoc (Nurkanović and Diehl, 2022).

Hamilton–Jacobi theory for nonholonomic and forced hybrid mechanical systems

A hybrid system is a system whose dynamics is given by a mixture of both continuous and discrete transitions. In particular, these systems can be utilized to describe the dynamics of a mechanical system with impacts. Based on the approach by Clark [Invariant measures, geometry and control of hybrid and nonholonomic dynamical systems, Ph.D thesis, University of Mirchigan (2020)], we develop a geometric Hamilton–Jacobi theory for forced and nonholonomic hybrid dynamical systems. We state the corresponding Hamilton–Jacobi equations for these classes of systems and apply our results to analyze some examples.

Enhancing control of dynamic flow systems through innovator controller design and parametric polynomial modeling

Enhancing control of dynamic flow systems through innovator controller design and parametric polynomial modeling

Robust adaptive control of uncertain dynamic systems using self-evolving neural learning technique

Robust adaptive control of uncertain dynamic systems using self-evolving neural learning technique

Editorial Conclusion for the Special Issue “New Theory and Applications of Nonlinear Analysis, Fractional Calculus and Optimization”

Nonlinear analysis has widespread and significant applications in many areas at the core of many branches of pure and applied mathematics and modern science, including nonlinear ordinary and partial differential equations, critical point theory, functional analysis, fixed point theory, nonlinear optimization, fractional calculus, variational analysis, convex analysis, dynamical system theory, mathematical economics, data mining, signal processing, control theory, and many more [...]

Enhancing the trustworthiness of chaos and synchronization of chaotic satellite model: a practice of discrete fractional-order approaches

Accurate development of satellite maneuvers necessitates a broad orbital dynamical system and efficient nonlinear control techniques. For achieving the intended formation, a framework of a discrete fractional difference satellite model is constructed by the use of commensurate and non-commensurate orders for the control and synchronization of fractional-order chaotic satellite system. The efficacy of the suggested framework is evaluated employing a numerical simulation of the concerning dynamic systems of motion while taking into account multiple considerations such as Lyapunov exponent research, phase images and bifurcation schematics. With the aid of discrete nabla operators, we monitor the qualitative behavioural patterns of satellite systems in order to provide justification for the structure’s chaos. We acquire the fixed points of the proposed trajectory. At each fixed point, we calculate the eigenvalue of the satellite system’s Jacobian matrix and check for zones of instability. The outcomes exhibit a wide range of multifaceted behaviours resulting from the interaction with various fractional-orders in the offered system. Additionally, the sample entropy evaluation is employed in the research to determine complexities and endorse the existence of chaos. To maintain stability and synchronize the system, nonlinear controllers are additionally provided. The study highlights the technique’s vulnerability to fractional-order factors, resulting in exclusive, changing trends and equilibrium frameworks. Because of its diverse and convoluted behaviour, the satellite chaotic model is an intriguing and crucial subject for research.

Adaptive dynamic programming and distributionally robust optimal control of linear stochastic system using the Wasserstein metric

SummaryIn this paper, we consider the optimal control of unknown stochastic dynamical system for both the finite‐horizon and infinite‐horizon cases. The objective of this paper is to find an optimal controller to minimize the expected value of a function which depends on the random disturbance. Throughout this paper, it is assumed that the mean vector and covariance matrix of the disturbance distribution is unknown. An uncertainty set in the space of mean vector and the covariance matrix is introduced. For the finite‐horizon case, we derive a closed‐form expression of the unique optimal policy and the opponents policy that generates the worst‐case distribution. For the infinite‐horizon case, we simplify the Riccati equation obtained in the finite‐hozion setting to an algebraic Riccati equation, which can guarantee the existence of the solution of the Riccati equation. It is shown that the resulting optimal policies obtained in these two cases can stabilize the expected value of the system state under the worst‐case distribution. Furthermore, the unknown system matrices can also be explicitly computed using the adaptive dynamic programming technique, which can help compute the optimal control policy by solving the algebraic Riccati equation. Finally, a simulation example is presented to demonstrate the effectiveness of our theoretical results.

Artificial neural network‐based adaptive control for nonlinear dynamical systems

SummaryThis research article presents an artificial neural network (ANN)‐based indirect adaptive control method for nonlinear dynamical systems. In this article, a modified Elman recurrent neural network (MERNN) is proposed as an identifier and controller for controlling nonlinear systems. The architecture of the proposed controller is a modified form of the existing Elman recurrent neural network. The parameter training of ANN‐based controllers is obtained by using the most popular optimization algorithm which is known as the back‐propagation algorithm. A comparative study includes Elman, Diagonal, Jordan, feed‐forward neural network (FFNN), and radial basis function network (RBFN)‐based controllers to compare with the proposed MERNN controller. To determine the controller's robustness, parameter variations, and disturbance signals have been considered. The performance analysis of the proposed controller is illustrated by two simulation examples. The simulation results reveal that MERNN can not only identify the unknown dynamics of the plant but also adaptively control it compared to the others.

Reliability control of nonlinear stochastic dynamical system based on discrete data

Stochastic optimal control based on reliability plays a crucial role in mitigating structural failure and ensuring safe operation. For general stochastic vibration systems, a data-driven method for reliability-based optimal control is proposed based on random state data. Firstly, the feedback control is split into conservative and dissipative components in coherence with physical intuition, and each component is expanded upon using pre-selected basic functions, resulting in a modified system with undetermined coefficients. Then, from discrete random samples, the expressions for the probability densities of the first-passage time and the stationary response can be identified, explicitly including system and excitation parameters, as well as initial and safety boundary conditions. Finally, a performance index, combining system reliability and control cost, is constructed and can be explicitly expressed in terms of the undetermined coefficients of the control forces. The optimal control problem for determining the optimal control forces is reformulated as an optimization problem for determining the coefficients that minimize the performance index. This unconstrained optimization problem can be easily solved. Three typical nonlinear stochastic systems, a controlled Duffing oscillator, a controlled nonlinear hysteretic system, and a controlled 2-DOF nonlinear dynamical system, are given to illustrate the validity and accuracy of the proposed data-driven method. Additionally, the control performance of the optimal control strategy in improving the system reliability is also discussed.

Hermite kernel surrogates for the value function of high-dimensional nonlinear optimal control problems

Numerical methods for the optimal feedback control of high-dimensional dynamical systems typically suffer from the curse of dimensionality. In the current presentation, we devise a mesh-free data-based approximation method for the value function of optimal control problems, which partially mitigates the dimensionality problem. The method is based on a greedy Hermite kernel interpolation scheme and incorporates context knowledge by its structure. Especially, the value function surrogate is elegantly enforced to be 0 in the target state, non-negative and constructed as a correction of a linearized model. The algorithm allows formulation in a matrix-free way which ensures efficient offline and online evaluation of the surrogate, circumventing the large-matrix problem for multivariate Hermite interpolation. Additionally, an incremental Cholesky factorization is utilized in the offline generation of the surrogate. For finite time horizons, both convergence of the surrogate to the value function and for the surrogate vs. the optimal controlled dynamical system are proven. Experiments support the effectiveness of the scheme, using among others a new academic model with an explicitly given value function. It may also be useful for the community to validate other optimal control approaches.

Optimal feedback control of dynamical systems via value-function approximation

Optimal feedback control of dynamical systems via value-function approximation

Adaptive control for hybrid dynamical systems with user-defined rate of convergence

This paper presents a model reference adaptive control (MRAC) system for nonlinear, time-varying, hybrid dynamical plants affected by modeling uncertainties. This system is unique because, in addition to regulating the trajectory tracking error to zero, it allows the user to set the rate of convergence of the tracking error between consecutive resetting events. This proposed result has been possible by leveraging an auxiliary trajectory tracking error dynamics to accelerate convergence of the trajectory tracking error according to the two-layer MRAC framework. The resetting logic of both the reference model and the auxiliary trajectory tracking error dynamics bound variations in the trajectory tracking error due to undesired resetting events in the plant dynamics. The resetting events of the reference and auxiliary model also reduce the trajectory tracking error whenever some dwell-time condition is met. Finally, the resetting logic of both the reference model and the auxiliary trajectory tracking error reduces the control effort at isolated time-instants. Three numerical examples demonstrate the applicability of the proposed result as well as their advantages over alternative adaptive control techniques for nonlinear, time-varying, hybrid dynamical plants as well as classical MRAC.

Приближенный синтез оптимальных детерминированных систем управления с неполной обратной связью на основе достаточных условий ε-оптимальности

<p>The problem of optimal control of deterministic dynamical systems in the absence of information about a part of the coordinates of the state vector is considered. Sufficient ε-optimality conditions based on the principle of expansion are formulated and proved. An algorithm is proposed for finding an a priori estimate of the proximity of the synthesized control law with incomplete feedback to the optimal one for a given set of initial states. The solution of the model example is given.</p>

Physics-Informed Neural Networks with skip connections for modeling and control of gas-lifted oil wells

Neural networks, while powerful, often lack interpretability. Physics-Informed Neural Networks (PINNs) address this limitation by incorporating physics laws into the loss function, making them applicable to solving Ordinary Differential Equations (ODEs) and Partial Differential Equations (PDEs). The recently introduced PINC framework extends PINNs to control applications, allowing for open-ended long-range prediction and control of dynamic systems. In this work, we enhance PINC for modeling highly nonlinear systems such as gas-lifted oil wells. By introducing skip connections in the PINC network and refining certain terms in the ODE, we achieve more accurate gradients during training, resulting in an effective modeling process for the oil well system. Our proposed improved PINC demonstrates superior performance, reducing the validation prediction error by an average of 67% in the oil well application and significantly enhancing gradient flow through the network layers, increasing its magnitude by four orders of magnitude compared to the original PINC. Furthermore, experiments showcase the efficacy of Model Predictive Control (MPC) in regulating the bottom-hole pressure of the oil well using the improved PINC model, even in the presence of noisy measurements.

Data-Driven Stochastic Optimal Control With Safety Constraints Using Linear Transfer Operators

We provide a data-driven framework for optimal control of a continuous-time control-affine stochastic dynamical system. The proposed framework relies on the linear operator theory involving Perron-Frobenius (P-F) and Koopman operators. Our first result involving the P-F operator provides a convex formulation to the optimal control problem in the dual space of densities. This convex formulation of the stochastic optimal control problem leads to an infinite-dimensional convex program. The finite-dimensional approximation of the convex program is obtained using a data-driven approximation of the linear operators. We provide rigorous error bounds and convergence rate for the data-driven approximation of the linear operators and comment on the convergence rate of the optimal control w.r.t. data size. Our second results demonstrate using the Koopman operator, dual to the P-F operator, for the stochastic optimal control design. We show that the Hamilton Jacobi Bellman (HJB) equation can be expressed using the Koopman operator. We provide an iterative procedure along the lines of a popular policy iteration algorithm based on the data-driven approximation of the Koopman operator for solving the HJB equation. The two formulations, namely the convex formulation involving the P-F operator and the Koopman-based formulation using the HJB equation, can be viewed as dual to each other where the duality follows due to the dual nature of the P-F and Koopman operators. Finally, we present examples to demonstrate the efficacy of the developed framework and verify the convergence rates for the operator and optimal control numerically.

Non-Quadratic Proxy Functions in Mirror Descent Method Applied to Designing of Robust Controllers for Nonlinear Dynamic Systems with Uncertainty

Non-Quadratic Proxy Functions in Mirror Descent Method Applied to Designing of Robust Controllers for Nonlinear Dynamic Systems with Uncertainty

On the application of the calculus of positively constructed formulas for the study of controlled discrete-event systems

The article is devoted to the development of an approach to solving the main problems of the theory of supervisory control of logical discrete-event systems (DES), based on their representation in the form of positively constructed formulas (PCF). We consider logical DESs in automata form, understood as generators of some regular languages. The PCF language is a complete first-order language, the formulas of which have a regular structure of alternating type quantifiers and do not contain a negation operator in the syntax. It was previously proven that any formula of the classical first-order predicate calculus can be represented as a PCF. PCFs have a visual tree representation and a natural question-and-answer procedure for searching for an inference using a single inference rule. It is shown how the PCF calculus, developed in the 1990s to solve some problems of control of dynamic systems, makes it possible to solve basic problems of the theory of supervisory control, such as checking the criteria for the existence of supervisory control, automatically modifying restrictions on the behavior of the controlled system, and implementing a supervisor. Due to some features of the PCF calculus, it is possible to use a non-monotonic inference. It is demonstrated how the presented PCF-based method allows for additional event processing during inference. The Bootfrost software system, or the so-called prover, designed to refute the obtained PCFs is also presented, and the features of its implementation are briefly described. As an illustrative example, we consider the problem of controlling an autonomous mobile robot.

Stability Analysis of Switched Linear Systems with Neural Lyapunov Functions

Neural-based, data-driven analysis and control of dynamical systems have been recently investigated and have shown great promise, e.g. for safety verification or stability analysis. Indeed, not only do neural networks allow for an entirely model-free, data-driven approach, but also for handling arbitrary complex functions via their power of representation (as opposed to, e.g. algebraic optimization techniques that are restricted to polynomial functions). Whilst classical Lyapunov techniques allow to provide a formal and robust guarantee of stability of a switched dynamical system, very little is yet known about correctness guarantees for Neural Lyapunov functions, nor about their performance (amount of data needed for a certain accuracy). We formally introduce Neural Lyapunov functions for the stability analysis of switched linear systems: we benchmark them on this paradigmatic problem, which is notoriously difficult (and in general Turing-undecidable), but which admits existing recently-developed technologies and theoretical results. Inspired by switched systems theory, we provide theoretical guarantees on the representative power of neural networks, leveraging recent results from the ML community. We additionally experimentally display how Neural Lyapunov functions compete with state-of-the-art results and techniques, while admitting a wide range of improvement, both in theory and in practice. This study intends to improve our understanding of the opportunities and current limitations of neural-based data-driven analysis and control of complex dynamical systems.

Design of Car Control Algorithm Based on Deep Reinforcement Learning

Reinforcement learning algorithm is widely used in the field of agent control because of its high robustness and strong adaptability. The problem of Mountain-Car-v0 is a classic control problem, which attracts the attention of many researchers. Traditional control methods, for example, PID control, are widely used in industrial and robot control fields, but they show obvious limitations in the face of such highly nonlinear and unknown environments. Therefore, this paper proposes a car control algorithm based on deep reinforcement learning, aiming at overcoming the shortcomings of traditional methods and realizing efficient control of complex dynamic systems. By combining the automatic extraction of high-dimensional features of deep learning with the dynamic decision-making mechanism of reinforcement learning, an adaptive control algorithm is designed, which can continuously learn and optimize in the interaction with the environment without establishing an accurate system model in advance. The experimental results show that compared with traditional control methods and other intelligent control strategies, the proposed algorithm shows higher efficiency and better adaptability in the task of Mountain-car-V0, and obviously improves the success rate of the Mountain-Car-v0.