Parameter identification of linear time‐invariant systems using dynamic regressor extension and mixing

Dynamic regressor extension and mixing (DREM) is a new technique for parameter estimation that has proven instrumental in the solution of several open problems in system identification and adaptive control. A key property of the estimator is that, by generation of scalar regression models, it guarantees monotonicity of each element of the parameter error vector that is a much stronger property than monotonicity of the vector norm , as ensured with classical gradient or least‐squares estimators. On the other hand, the overall performance improvement of the estimator is strongly dependent on the suitable choice of certain operators that enter in the design. In this paper, we investigate the impact of these operators on the convergence properties of the estimator in the context of identification of linear single‐input single‐output time‐invariant systems with periodic excitation. The most important contribution is that the DREM (almost surely) converges under the same persistence of excitation (PE) conditions as the gradient estimator while providing improved transient performance. In particular, we give some guidelines how to select the DREM operators to ensure convergence under the same PE conditions as standard identification schemes.     

Robust adaptive control for unmatched systems with guaranteed parameter estimation convergence

This paper provides a modified model reference adaptive control (MRAC) scheme to achieve better transient control performance for systems with unknown unmatched dynamics, where an adaptive law with guaranteed convergence is introduced. We first revisit the standard MRAC system and analyze the tracking error bound by using L₂‐norm and Cauchy‐Schwartz inequality. Based on this analysis, we suggest a feasible way to compensate the undesired transient dynamics induced by the gradient descent–based adaptive laws subject to sluggish convergence or even parameter drift. Then, a modified adaptive law with an alternative leakage term containing the parameter estimation error is developed. With this adaptive law, the convergence of both the estimation error and tracking error can be proved simultaneously. This enhanced convergence property can contribute to deriving smoother control signal and improved control response. Moreover, this paper provides a simple and numerically feasible approach to online verify the well‐known persistent excitation condition by testing the positive definiteness of an introduced auxiliary matrix. Comparative simulations based on a benchmark 3‐DOF helicopter model are given to validate the effectiveness of the proposed MRAC approach and show the improved performance over several other MRAC schemes.     

Transient performance improvement in indirect model reference adaptive control using perturbation‐based extremum seeking identifier

One of the main drawbacks of model reference adaptive control (MRAC) is the weakness of its transient performance. The key reason of this imperfection is parameter's estimation error convergence. For many cases in the closed‐loop control, the plant input signal cannot satisfy the persistence of excitation (PE) condition which yields poor parameters estimation error convergence. In this paper, we use a fast perturbation‐based extremum seeking (PES) scheme without steady‐state oscillation as the parameter identifier in indirect MRAC. The estimated parameters through the PES identifier contain the additive sinusoidal signals with distinct frequencies in the transient, which satisfy the PE condition of the plant input. Therefore, convergence of the parameters estimation error to zero will be guaranteed that results in improvement of transient performance for indirect MRAC. Also, the contrary effects on the steady‐state behaviour is eliminated since the sinusoidal excitation signals amplitude exponentially converge to zero and reinitiate with every change in the unknown parameters. Simulation results for a second order example have been presented to illustrate the effectiveness of the proposed scheme.     

Experience replay–based output feedback Q‐learning scheme for optimal output tracking control of discrete‐time linear systems

This paper focuses on solving the adaptive optimal tracking control problem for discrete‐time linear systems with unknown system dynamics using output feedback. A Q‐learning‐based optimal adaptive control scheme is presented to learn the feedback and feedforward control parameters of the optimal tracking control law. The optimal feedback parameters are learned using the proposed output feedback Q‐learning Bellman equation, whereas the estimation of the optimal feedforward control parameters is achieved using an adaptive algorithm that guarantees convergence to zero of the tracking error. The proposed method has the advantage that it is not affected by the exploration noise bias problem and does not require a discounting factor, relieving the two bottlenecks in the past works in achieving stability guarantee and optimal asymptotic tracking. Furthermore, the proposed scheme employs the experience replay technique for data‐driven learning, which is data efficient and relaxes the persistence of excitation requirement in learning the feedback control parameters. It is shown that the learned feedback control parameters converge to the optimal solution of the Riccati equation and the feedforward control parameters converge to the solution of the Sylvester equation. Simulation studies on two practical systems have been carried out to show the effectiveness of the proposed scheme.     

A new model reference adaptive control for linear time‐varying systems

Recent results on the adaptive control of linear time‐varying systems have considered mostly the case in which the range or rate of parameter variations is small. In this paper, a new state feed‐back model reference adaptive control is developed for systems with bounded arbitrary parameter variations. The important feature of the proposed adaptive control is an uncertainty estimation algorithm, which guarantees almost zero tracking error. Note that the conventional parameter estimation algorithm in the adaptive control guarantees only bounded tracking error. Copyright © 2000 John Wiley & Sons, Ltd.     

Indirect adaptive control of a class of marine vehicles

A nonlinear adaptive framework for bounded‐error tracking control of a class of non‐minimum phase marine vehicles is presented. The control algorithm relies on a special set of tracking errors to achieve satisfactory tracking performance while guaranteeing stable internal dynamics. First, the design of a model‐based nonlinear control law, guaranteeing asymptotic stability of the error dynamics, is presented. This control algorithm solves the tracking problem for the considered class of marine vehicles, assuming full knowledge of the system model. Then, the analysis of the zero‐dynamics is carried out, which illustrates the efficacy of the chosen set of tracking errors in stabilizing the internal dynamics. Finally, an indirect adaptive technique, relying on a partial state predictor, is used to address parametric uncertainties in the model. The resulting adaptive control algorithm guarantees Lyapunov stability of the errors and parameter estimates, as well as asymptotic convergence of the errors to zero. Numerical simulations illustrate the performance of the adaptive algorithm. Copyright © 2009 John Wiley & Sons, Ltd.     

Composite learning fuzzy synchronization for incommensurate fractional‐order chaotic systems with time‐varying delays

This paper presents a composite learning fuzzy control to synchronize two different uncertain incommensurate fractional‐order time‐varying delayed chaotic systems with unknown external disturbances and mismatched parametric uncertainties via the Takagi‐Sugeno fuzzy method. An adaptive controller together with fractional‐order composite learning laws is designed based on both a parallel distributed compensation technology and a fractional Lyapunov criterion. The boundedness of all variables in the closed‐loop system and the Mittag‐Leffler stability of tracking error can be guaranteed. T‐S fuzzy systems are provided to tackle unknown nonlinear functions. The distinctive features of the proposed approach consist in the following: (1) a supervisory control law is designed to compensate the lumped disturbances; (2) both the prediction error and the tracking error are used to estimate the unknown fuzzy system parameters; (3) parameter convergence can be ensured by an interval excitation condition. Finally, the feasibility of the proposed control strategy is demonstrated throughout an illustrative example.     

Adaptive tracking control of Euler–Lagrange systems with bounded controls

We solve the simultaneous closed‐loop identification and tracking‐control problems for fully‐actuated Euler–Lagrange systems under input constraints. We use a nonlinear adaptive controller reminiscent of computed‐torque‐type controllers in which linear correction terms are saturated in order to comply with the imposed bounds on the control inputs. Adaptation, reminiscent of gradient methods, is used also with saturation. With respect to related literature, our contribution consists in establishing uniform global asymptotic stability. Therefore, our control scheme ensures robustness with respect to bounded perturbations and uniform convergence of the estimation errors for any initial conditions. Copyright © 2016 John Wiley & Sons, Ltd.     

Inverse optimal adaptive control for non‐linear uncertain systems with exogenous disturbances

A Lyapunov‐based inverse optimal adaptive control‐system design problem for non‐linear uncertain systems with exogenous ℒ︁₂ disturbances is considered. Specifically, an inverse optimal adaptive non‐linear control framework is developed to explicitly characterize globally stabilizing disturbance rejection adaptive controllers that minimize a nonlinear‐nonquadratic performance functional for non‐linear cascade and block cascade systems with parametric uncertainty. It is shown that the adaptive Lyapunov function guaranteeing closed‐loop stability is a solution to the Hamilton–Jacobi–Isaacs equation for the controlled system and thus guarantees both optimality and robust stability. Additionally, the adaptive Lyapunov function is dissipative with respect to a weighted input–output energy supply rate guaranteeing closed‐loop disturbance rejection. For special integrand structures of the performance functionals considered, the proposed adaptive controllers additionally guarantee robustness to multiplicative input uncertainty. In the case of linear‐quadratic control it is shown that the operations of parameter estimation and controller design are coupled illustrating the breakdown of the certainty equivalence principle for the optimal adaptive control problem. Finally, the proposed framework is used to design adaptive controllers for jet engine compression systems with uncertain system dynamics. Copyright © 2000 John Wiley & Sons, Ltd.     

Adaptive set‐point regulation of discrete‐time nonlinear systems

In this paper, adaptive set‐point regulation controllers for discrete‐time nonlinear systems are constructed. The system to be controlled is assumed to have a parametric uncertainty, and an excitation signal is used in order to obtain the parameter estimate. The proposed controller belongs to the category of indirect adaptive controllers, and its construction is based on the policy of calculating the control input rather than that of obtaining a control law. The proposed method solves the adaptive set‐point regulation problem under the assumption that the target state is reachable for each fixed parameter value. Additional feature of the proposed method is that Lyapunov‐like functions have not been used in the construction of the controllers. Copyright © 2015 John Wiley & Sons, Ltd.     

New solutions to the 2D adaptive visual servoing problem with relaxed excitation requirements

In this paper, we present three new globally convergent vision‐based position controllers for a planar manipulator in a fixed‐camera configuration, where the camera orientation and scale factor are considered unknown. This is a basic adaptive visual servoing problem whose solution was hampered by the nonlinear dependence of the system dynamics on the unknown parameters. Proposing a suitable reparameterization of the systems mathematical model, and exploiting some structural properties of it, we propose three different solutions to the problem. The first one is the certainty equivalent version of the known parameter controller and requires some excitation conditions to ensure global asymptotic convergence. A second version of the controller, which is now slightly more complicated and, possibly, needs to inject some high gain but requires significantly weaker excitation conditions, is given. Finally, we propose a slight modification to the second scheme to achieve the trajectory tracking in finite time. The efficacy of the three adaptive controllers is shown through realistic simulations.     

Concurrent learning adaptive control of linear systems with exponentially convergent bounds

Concurrent learning adaptive controllers, which use recorded and current data concurrently for adaptation, are developed for model reference adaptive control of uncertain linear dynamical systems. We show that a verifiable condition on the linear independence of the recorded data is sufficient to guarantee global exponential stability. We use this fact to develop exponentially decaying bounds on the tracking error and weight error, and estimate upper bounds on the control signal. These results allow the development of adaptive controllers that ensure good tracking without relying on high adaptation gains, and can be designed to avoid actuator saturation. Simulations and hardware experiments show improved performance. Copyright © 2012 John Wiley & Sons, Ltd.     

Open‐loop Stackelberg learning solution for hierarchical control problems

This work presents a novel framework based on adaptive learning techniques to solve the continuous‐time open‐loop Stackelberg games. The method yields real‐time approximations of the game value and convergence of the policies to the open‐loop Stackelberg‐equilibrium solution, while also guaranteeing asymptotic stability of the equilibrium point of the closed‐loop system. It is implemented as a separate actor/critic parametric network approximator structure for every player and involves simultaneous continuous‐time adaptation. To introduce and implement the hierarchical structure to the coupled optimization problem, we adjoin to the leader the controller dynamics of the follower. A persistence of excitation condition guarantees convergence of both critics to the actual game values that eventually solve the hierarchical optimization problem. A simulation example shows the efficacy of the proposed approach.     

Learning‐based iterative modular adaptive control for nonlinear systems

In this paper, we study the problem of adaptive trajectory tracking control for a class of nonlinear systems with structured parametric uncertainties. We propose to use an iterative modular approach: we first design a robust nonlinear state feedback that renders the closed‐loop input‐to‐state stable (ISS). Here, the input is considered to be the estimation error of the uncertain parameters, and the state is considered to be the closed‐loop output tracking error. Next, we propose an iterative adaptive algorithm, where we augment this robust ISS controller with an iterative data‐driven learning algorithm to estimate online the parametric uncertainties of the model. We implement this method with two different learning approaches. The first one is a data‐driven multiparametric extremum seeking method, which guarantees local convergence results, and the second is a Bayesian optimization‐based method called Gaussian Process Upper Confidence Bound, which guarantees global results in a compact search set. The combination of the ISS feedback and the data‐driven learning algorithms gives a learning‐based modular indirect adaptive controller. We show the efficiency of this approach on a two‐link robot manipulator numerical example.     

Real-time implementation of an indirect adaptive robot controller

Adaptive controllers applied to high-speed and high-precision robot manipulators give excellent tracking performances because they take into account the full dynamics of the robot. Recent publications have shown that on-line estimation of the dynamic constant parameters can be obtained by the joint tracking error (direct adaptive control) or by the torque prediction error (indirect adaptive control). In direct adaptive controllers the estimation law is derived from Lyapunov stability or Popov hyperstability methods. These controllers are simple and their real-time implementation is easy; however, the estimation is not so accurate. In indirect adaptive controllers the estimation law is based on a least squares algorithm. These controllers give accurate estimates of the manipulator parameters; however, they involve much more computation than the direct approach. Therefore no real-time implementations of indirect adaptive tracking controllers for robots have been reported in the literature until now. This paper describes a real-time implementation of an indirect adaptive scheme applied to a two-degree-of-freedom (2DOF) direct-drive SCAM robot. the controller is implemented at low cost by the use of a single-chip digital signal processor (DSP).     

Adaptive neural control for nonstrict‐feedback stochastic nonlinear time‐delay systems with input and output constraints

This paper investigates an adaptive neural tracking control for a class of nonstrict‐feedback stochastic nonlinear time‐delay systems with input saturation and output constraint. First, the Gaussian error function is used to represent a continuous differentiable asymmetric saturation model. Second, the appropriate Lyapunov‐Krasovskii functional and the property of hyperbolic tangent functions are used to compensate the time‐delay effects, the neural network is used to approximate the unknown nonlinearities, and a barrier Lyapunov function is designed to ensure that the output parameters are restricted. At last, based on Lyapunov stability theory, a robust adaptive neural control method is proposed, and the designed controller decreases the number of learning parameters and thus reduces the computational burden. It is shown that the designed neural controller can ensure that all the signals in the closed‐loop system are 4‐Moment (or 2 Moment) semi‐globally uniformly ultimately bounded and the tracking error converges to a small neighborhood of the origin. Two examples are given to further verify the effectiveness of the proposed approach.     

Robustness of linear time‐varying systems with relaxed excitation

It is a well‐known fact that linear time‐varying systems with a persistently excited state matrix are exponentially converging and input‐to‐state stable with respect to additive perturbations. Recently, several relaxed conditions of persistent excitation have been presented, which ensure an asymptotic convergence rate in the system. In the present work, it is shown that these conditions are similar and that, under such a relaxed excitation, only nonuniform in time input‐to‐state stability and integral input‐to‐state stability properties can be obtained. The results are illustrated by simulations for a problem of estimation in the linear regression model.     

Control of robots using radial basis function neural networks with dead‐zone

In this paper, we examine the control of robot manipulators utilizing a Radial Basis Function (RBF) neural network. We are able to remove the typical requirement of Persistence of Excitation (PE) for the desired trajectory by introducing an error minimizing dead‐zone in the learning dynamics of the neural network. The dead‐zone freezes the evolution of the RBF weights when the performance error is within a bounded region about the origin. This guarantees that the weights do not go unbounded even if the PE condition is not imposed. Utilizing protection ellipsoids we derive conditions on the feedback gain matrices that guarantee that the origin of the closed loop system is semi‐globally uniformly bounded. Simulations are provided illustrating the techniques. Copyright © 2011 John Wiley & Sons, Ltd.     

Adaptive robust dynamic surface control with composite adaptation laws

This paper focuses on an adaptive robust dynamic surface control (ARDSC) with composite adaptation laws (CAL) for a class of uncertain nonlinear systems in semi‐strict feedback form. A simple and effective controller has been obtained by introducing dynamic surface control (DSC) technique and designing novel adaptation laws. First, the ‘explosion of terms’ problem caused by backstepping method in the traditional adaptive robust control (ARC) is avoided. Meanwhile, through a new proof philosophy the asymptotical output tracking that the ARC possesses is theoretically preserved. Second, when persistent excitation (PE) condition satisfies, true parameter estimates could be acquired via designing CALs which integrate the information of estimation errors. Finally, simulation results are presented to illustrate the effectiveness of the proposed method. Copyright © 2010 John Wiley & Sons, Ltd.     

Fault‐tolerant control using command‐filtered adaptive back‐stepping technique: Application to hypersonic longitudinal flight dynamics

In this paper, the adaptive back‐stepping controller is investigated for a class of strict‐feedback systems using the command filter technique. Adaptive laws are designed for updating the controller parameters when both the plant parameters and actuator‐failure parameters are unknown. Furthermore, the auxiliary dynamics is developed to deal with the input constraints. Closed‐loop stability and asymptotic‐state tracking are ensured. The method is applied to the longitudinal dynamics of a generic hypersonic aircraft in the presence of actuator faults and input constraints. Based on the parameter estimation, the command‐filtered adaptive back‐stepping control is presented. Simulation results on the control‐oriented model show that the proposed approach achieves good tracking performance. Copyright © 2015 John Wiley & Sons, Ltd.