Adaptive pseudospectral methods for solving constrained linear and nonlinear time-delay optimal control problems

In this paper, we first develop an adaptive shifted Legendre–Gauss (ShLG) pseudospectral method for solving constrained linear time-delay optimal control problems. The delays in the problems are on the state and/or on the control input. By dividing the domain of the problem into a uniform mesh based on the delay terms, the constrained linear time-delay optimal control problem is reduced to a quadratic programming problem. Next, we extend the application of the adaptive ShLG pseudospectral method to nonlinear problems through quasilinearization. Using this scheme, the constrained nonlinear time-delay optimal control problem is replaced with a sequence of constrained linear-quadratic sub-problems whose solutions converge to the solution of the original nonlinear problem. The method is called the iterative-adaptive ShLG pseudospectral method. One of the most important advantages of the proposed method lies in the case with which nonsmooth optimal controls can be computed when inequality constraints and terminal constraints on the state vector are imposed. Moreover, a comparison is made with optimal solutions obtained analytically and/or other numerical methods in the literature to demonstrate the applicability and accuracy of the proposed methods.     

A hybrid approximation scheme for discretizing constrained quadratic optimal control problems

In this paper, a composite Chebyshev finite difference method for solving linear quadratic optimal control problems with inequality constraints on state and control variables is introduced. This method is an extension of Chebyshev finite difference scheme and is based on a hybrid of block-pulse functions and Chebyshev polynomials using the well known Chebyshev–Gauss–Lobatto nodes. The excellent properties of hybrid functions are used to convert optimal control problem into a mathematical programming problem whose solution is much more easier than the original one. Various types of optimal control problems are investigated to demonstrate the effectiveness of the proposed approximation scheme. The method is simple, easy to implement and provides very accurate results.     

Backstepping-based adaptive dynamic programming for missile-target guidance systems with state and input constraints

In this paper, a novel backstepping-based adaptive dynamic programming (ADP) method is developed to solve the problem of intercepting a maneuver target in the presence of full-state and input constraints. To address state constraints, a barrier Lyapunov function is introduced to every backstepping procedure. An auxiliary design system is employed to compensate the input constraints. Then, an adaptive backstepping feedforward control strategy is designed, by which the tracking problem for strict-feedback systems can be reduced to an equivalence optimal regulation problem for affine nonlinear systems. Secondly, an adaptive optimal controller is developed by using ADP technique, in which a critic network is constructed to approximate the solution of the associated Hamilton–Jacobi–Bellman (HJB) equation. Therefore, the whole control scheme consists of an adaptive feedforward controller and an optimal feedback controller. By utilizing Lyapunov's direct method, all signals in the closed-loop system are guaranteed to be uniformly ultimately bounded (UUB). Finally, the effectiveness of the proposed strategy is demonstrated by using a simple nonlinear system and a nonlinear two-dimensional missile-target interception system.     

Time-optimal control problem for a linear parameter varying system with nonlinear item

In this paper, we considered a time-optimal control problem for a new type of linear parameter varying (LPV) system which is obtained through data identification in the process of dealing with actual problems. The addition of non-linear terms is compensation for the method that does not require linear expansion at the equilibrium point. Since the objective function is the terminal time which is an implicit function concerning decision variables, it is a non-standard optimal control problem with uncertain terminal time. To find the global optimal solution to this problem, firstly, the control parameterization method is used to transform it into a nonlinear optimization problem of parameter selection, and then the modifed particle swarm optimization (PSO) algorithm is combined to solve the equivalent nonlinear programming problem. Numerical examples are used to illustrate the effectiveness of the proposed algorithm.     

Time-optimal Control Systems with Norm Constraints on State Variables at Intermediate Times

The problem of time-optimal control systems with both norm constraints on control inputs and on state variables at discrete intermediate times is formulated as an L-problem in the theory of moments. The simplex method is used for solving a nonlinear minimizing problem inherent in the functional analysis solution to this latter problem. Numerical results are presented for a train operation.     

Minimax optimal control of linear system with input-dependent uncertainty

In this paper, the quadratic minimax optimal control of linear system with input-dependent uncertainty is studied. We show that it admits a unique solution and can be approximated by a sequence of finite-dimensional minimax optimal parameter selection problems. These finite-dimensional minimax optimal parameter selection problems are further reduced to scalar optimization problems which also admit unique solutions. Thus, the original minimax optimal control problem is solved via solving a sequence of simple scalar optimization problems. Numerical experiments are presented to illustrate the developed method.     

A direct approach using the finite element method for the solution of the linear optimal control problem with a quadratic criterion

The present paper proposes a numerical approach to a linear optimal control problem with a quadratic performance index. In this technique, the time interval is divided into a number of time segments and all of the unknown functions which appear in the performance index are either interpolated linearly with respect to time or assumed to be constant in each time segment. The augmented performance index is discretized within each time element through the ordinary finite element technique.The main advantage of the present method is as follows: all of the necessary conditions for the performance index to be stationary can be expressed in the form of algebraic equations and the performance sequence of the state variables can be eliminated. As a result, the optimal control problem is reduced to the simple one of finding the sequence of control variables alone, which minimizes the quadratic performance index.A general formulation of the method is given and simple numerical examples are shown to demonstrate the effectiveness of the technique.     

A direct method based on the Chebyshev polynomials for a new class of nonlinear variable-order fractional 2D optimal control problems

The main goal of this study is to develop an efficient matrix approach for a new class of nonlinear 2D optimal control problems (OCPs) affected by variable-order fractional dynamical systems. The offered approach is established upon the shifted Chebyshev polynomials (SCPs) and their operational matrices. Through the way, a new operational matrix (OM) of variable-order fractional derivative is derived for the mentioned polynomials.The necessary optimality conditions are reduced to algebraic systems of equations by using the SCPs expansions of the state and control variables, and applying the method of constrained extrema. More precisely, the state and control variables are expanded in components of the SCPs with undetermined coefficients. Then these expansions are substituted in the cost functional and the 2D Gauss-Legendre quadrature rule is utilized to compute the double integral and consequently achieve a nonlinear algebraic equation.After that, the generated OM is employed to extract some algebraic equations from the approximated fractional dynamical system. Finally, the procedure of the constrained extremum is used by coupling the algebraic constraints yielded from the dynamical system and the initial and boundary conditions with the algebraic equation extracted from the cost functional by a set of unknown Lagrange multipliers. The method is established for three various types of boundary conditions.The precision of the proposed approach is examined through various types of test examples.Numerical simulations confirm the suggested approach is very accurate to provide satisfactory results.     

Adaptive neural dynamic surface control for full state constrained stochastic nonlinear systems with unmodeled dynamics

This paper solves the problem of adaptive neural dynamic surface control (DSC) for a class of full state constrained stochastic nonlinear systems with unmodeled dynamics. The concept of the state constraints in probability is first proposed and applied to the stability analysis of the system. The full state constrained stochastic nonlinear system is transformed to the system without state constraints through a nonlinear mapping. The unmodeled dynamics is dealt with by introducing a dynamic signal and the adaptive neural dynamic surface control method is explored for the transformed system. It is proved that all signals of the closed-loop system are bounded in probability and the error signals are semi-globally uniformly ultimately bounded(SGUUB) in mean square or the sense of four-moment. At the same time, the full state constraints are not violated in probability. The validity of the proposed control scheme is demonstrated through the simulation examples.     

区间两相模糊多目标模型在种植结构优化中的应用——以辽宁省大连市为例

针对基于“Max-min”算子的区间模糊多目标规划仅采用一或两个控制变量放松所有目标和模糊约束会造成某些约束过满意而某些约束不满意的情况,本文引入两相模糊规划,构建了区间-两相模糊多目标规划模型,并以辽宁省大连市种植结构优化为例进行研究。结果表明,该模型引入多个控制变量放松每个不确定目标和约束条件,且要求它们分别不小于“Max-min”算子中相应目标和约束条件的隶属度,更充分地利用了约束资源,保证了求解的有效性,减少了农业灌溉用水量;另外区间形式的最优解及4种不同情景的优化方案为决策者提供了决策空间,更真实地反映输入参数的不确定性对配置结果的影响。     

A novel mesh discretization strategy for numerical solution of optimal control problems in aerospace engineering

An effective approach is proposed for optimal control problems in aerospace engineering. First, several interval lengths are treated as optimization variables directly to localize the switching points accurately. Second, the variable intervals are usually refined into more subintervals homogeneously to obtain the trajectories with high accuracy. To reduce the number of optimization variables and improve the efficiency, the control and the state vectors are parameterized using different meshes in this paper such that the control can be approximated asynchronously by fewer parameters where the trajectories change slowly. Then, the variables are departed as independent variables and dependent variables, the gradient formulae, based on the partial derivatives of dependent parameters with respect to independent parameters, are computed to solve nonlinear programming problems. Finally, the proposed approach is applied to the classic moon lander and hang glider problems. For the moon lander problem, the proposed approach is compared with CVP, Fast-CVP and GPM methods, respectively. For the hang glider problem, the proposed approach is compared with trapezoidal discretization and stopping criteria methods, respectively. The numerical results validate the effectiveness of the proposed approach.     

Permissible control of general constrained mechanical systems

This paper develops a unified approach for modeling and controlling mechanical systems that are constrained with general holonomic and nonholonomic constraints. The approach conceptually distinguishes and separates constraints that are imposed on the mechanical system for developing its physical structure between constraints that may be used for control purposes. This gives way to a general class of nonlinear control systems for constrained mechanical systems in which the control inputs are viewed as the permissible control forces. In light of this view, a new and simple technique for designing nonlinear state feedback controllers for constrained mechanical systems is presented. The general applicability of the approach is demonstrated by considering the nonlinear control of an underactuated system.     

General value iteration based single network approach for constrained optimal controller design of partially-unknown continuous-time nonlinear systems

In this paper, a novel iterative approximate dynamic programming scheme is proposed by introducing the learning mechanism of value iteration (VI) to solve the constrained optimal control problem for CT affine nonlinear systems with utilizing only one neural network. The idea is to show the feasibility of introducing the VI learning mechanism to solve for the constrained optimal control problem from a theoretical point of view, and thus the initial admissible control can be avoided compared with most existing works based on policy iteration (PI). Meanwhile, the initial condition of the proposed VI based method can be more general than the traditional VI method which requires the initial value function to be a zero function. A general analytical method is proposed to demonstrate the convergence property. To simplify the architecture, only one critic neural network is adopted to approximate the iterative value function while implementing the proposed method. At last, two simulation examples are proposed to validate the theoretical results.     

Disturbance observer-based robust missile autopilot design with full-state constraints via adaptive dynamic programming

This paper aims to develop a robust optimal control method for longitudinal dynamics of missile systems with full-state constraints suffering from mismatched disturbances by using adaptive dynamic programming (ADP) technique. First, the constrained states are mapped by smooth functions, thus, the considered systems become nonlinear systems without state constraints subject to unknown approximation error. In order to estimate the unknown disturbances, a nonlinear disturbance observer (NDO) is designed. Based on the output of disturbance observer, an integral sliding mode controller (ISMC) is derived to counteract the effects of disturbances and unknown approximation error, thus ensuring the stability of nonlinear systems. Subsequently, the ADP technique is utilized to learn an adaptive optimal controller for the nominal systems, in which a critic network is constructed with a novel weight update law. By utilizing the Lyapunov's method, the stability of the closed-loop system and the convergence of the estimation weight for critic network are guaranteed. Finally, the feasibility and effectiveness of the proposed controller are demonstrated by using longitudinal dynamics of a missile.     

Stable model predictive control for a nonlinear system

In this paper, a stable model predictive control approach is proposed for constrained highly nonlinear systems. The technique is a modification of the multistep Newton-type control strategy, which was introduced by Li and Biegler. The proposed control technique is applied on a constrained highly nonlinear aerodynamic test bed, the twin rotor MIMO system (TRMS) to show the efficacy of the control technique. Since the accuracy of the plant model is vital in MPC techniques, the nonlinear state space equations of the system are derived considering all possible effective components. The nonlinear model is adaptively linearized during the prediction horizon. The linearized models of the system are employed to form a linear quadratic objective function subject to a set of inequality constraints due to the system input/output limits. The stability of the control system is guaranteed using the terminal equality constraints technique. The satisfactory performance of the proposed control algorithm on the TRMS validates the effectiveness and the reliability of the approach.     

A new direct method based on the Chebyshev cardinal functions for variable-order fractional optimal control problems

In this paper, a new direct method based on the Chebyshev cardinal functions is proposed to solve a class of variable-order fractional optimal control problems (V-OFOCPs). To this end, a new operational matrix (OM) of variable-order (V-O) fractional derivative in the Caputo sense is derived for these basis functions and is used to obtain an approximate solution for the problem under study. In the proposed method, the state and the control variables are expanded in terms of the Chebyshev cardinal functions with unknown coefficients, at first. Then, the OM of V-O fractional derivative and some properties of the Chebyshev cardinal functions are employed to achieve a nonlinear algebraic equation corresponding to the performance index and a nonlinear system of algebraic equations corresponding to the dynamical system in terms of the unknown coefficients. Finally, the method of constrained extremum is applied, which consists of adjoining the constraint equations derived from the given dynamical system and the initial conditions to the performance index by a set of undetermined Lagrange multipliers. As a result, the necessary conditions of optimality are derived as a system of algebraic equations in the unknown coefficients of the state variable, control variable, and Lagrange multipliers. Furthermore, some numerical examples of different types are demonstrated with their approximate solutions for confirming the high accuracy and applicability of the proposed method.     

Event-triggered optimal interaction control of the MRM system under the complex multi-task constraints: Design and experiments

The main challenges of modular robot manipulators (MRMs) with the environmental constraints include the avoidance of catastrophic collision and the precious contacting in the whole interaction process. Consequently, an event-triggered optimal interaction control method of MRMs under the complex multi-task constraints is presented in this paper. Firstly, on the basis of the joint torque feedback (JTF) technique, the dynamic model of constrained MRM subsystem is established. Secondly, the sensorless-based decentralized nonlinear disturbance observer (NDOB) is proposed to detect and identify the sudden external collision for each joint. Then, the performance index function is improved to achieve the interaction control, which contains the fusion state variable function, the influence of external collision, the known model term, and the estimation of model uncertainties through the radial basis function neural network (RBFNN) identifier. Further, based on event-triggered mechanism and adaptive dynamic programming (ADP) algorithm, the approximate event-triggered optimal interaction control strategy is acquired by the critic neural network (NN). Next, the closed-loop MRM system is demonstrated to be uniformly ultimately bounded (UUB) through the Lyapunov stability theorem. Finally, the experiments are achieved effectively for each joint on the platform, such that the feasibility and universality of the proposed interaction control approach are testified by the experimental results.     

Fixed-time composite neural learning control of state-constrained nonlinear uncertain systems

This paper presents a fixed-time composite neural learning control scheme for nonlinear strict-feedback systems subject to unknown dynamics and state constraints. To address the problem of state constraints, a new unified universal barrier Lyapunov function is proposed to convert the constrained system into an unconstrained one. Taking the unconstrained system, a modified fixed-time convergence state predictor is explored, enabling the prediction error for compensating the neural adaptive law to be obtained and improving the learning ability of online neural networks (NNs). Without employing fractional power terms or a complicated switching strategy to build the control law, a new method of constructing a smooth fixed-time dynamic surface control scheme is proposed. This overcomes the potential singularity problem and the explosion of complexity often encountered in fixed-time back-stepping designs. The representative features of our design are threefold. First, it is free of the fractional power terms, yet offers fixed-time convergence. Second, it addresses the state constraint problem without requiring a feasibility check. Third, it constructs a new state-predictor and enhances the approximation accuracy of NNs. The stability of the proposed control scheme is analyzed using the Lyapunov technique. Simulation results are presented to illustrate the effectiveness of the proposed controller.     

On consensus control of nonlinear multiagent systems satisfying incremental quadratic constraints

In this paper, we investigate the problem of leaderless consensus control for the multiagent systems whose nonlinear dynamics satisfying incremental quadratic constraints. A distributed dynamic consensus protocol, decided by communication among neighboring agents, is presented to render nonlinear agent consensus with appropriate coupling weights. Next, an observer-based distributed protocol is considered to ensure consensus of nonlinear system without knowing full state information. Further, extensions to consensus strategies with nonlinear dynamics for the leader-following fashion are also addressed. By comparison to the traditional nonlinear consensus control methodologies, the proposed approach generalizes the Lipschitz nonlinearity as well as the combined nonlinearity of one-sided Lipschitz condition and quadratic inner-boundness condition towards a more generalized type of nonlinearity, which shows us a less conservative result in the Lyapunov proof. Finally, the numerical simulations for six agents are illustrated to show the feasibility and performance of the proposed control protocol with or without the presence of the observer.