Abstract:This article is concerned with a composite adaptive robust control (CARC) for non‐strict feedback nonlinear systems with model uncertainties including parametric uncertainties and uncertain nonlinearities. It is assumed that the model uncertainties of the nonlinear system are bounded but unknown. Firstly, prescribed tracking performance is achieved by the CARC control law even the unknown parameters are time‐varying. This is accomplished firstly by designing a discontinuous projection that confines the unknown parameters in an estimated region. Then the robustness of the closed‐loop system is guaranteed by leveraging the deterministic robust control philosophy. Also, a command filter has been designed to avoid computing complex derivatives of virtual control law in the backstepping procedure. Secondly, a composite adaptation law that intelligently integrates information of both tracking error and prediction error is constructed for improving the performance of controllers. The high adaptation gain of CARC leads to higher parameter convergence speed and smaller tracking errors. Moreover, if the actual parameters are within the design range, asymptotic stability can be achieved in presence of parametric uncertainties only. In addition, exponential convergence of tracking errors and parameters can be shown under certain conditions. Finally, numerical examples are made to validate the superiority and effectiveness of the proposed control scheme.

Adaptive Multi-Innovation Gradient Identification Algorithms for a Controlled Autoregressive Autoregressive Moving Average Model

Hierarchical Least-Squares Based Iterative Identification for Multivariable Systems with Moving Average Noises

Multivariable CAR-like System Identification with Multi-innovation Gradient and Least Squares Algorithms

Improve recursive identification using multi-iteration

A Composite Adaptive Robust Control for Nonlinear Systems with Model Uncertainties

Two-stage Gradient-based Iterative Estimation Methods for Controlled Autoregressive Systems Using the Measurement Data

Maximum likelihood based multi‐innovation stochastic gradient identification algorithms for bilinear stochastic systems with ARMA noise

Cauchy kernel correntropy‐based robust multi‐innovation identification method for the nonlinear exponential autoregressive model in non‐Gaussian environment

Additional fractional gradient descent identification algorithm based on multi-innovation principle for autoregressive exogenous models

Data Filtering-Based Multi-innovation Stochastic Gradient Algorithm for Nonlinear Output Error Autoregressive Systems

Identification of Multi-Innovation Stochastic Gradients with Maximum Likelihood Algorithm Based on Ship Maneuverability and Wave Peak Models

Hierarchical Gradient Based Iterative Parameter Estimation Algorithm for Multivariable Output Error Moving Average Systems

Online learning of ARIMA for time series prediction.(2016)

Convergence of Recursive Identification for ARMAX Process with Increasing Variances

Three‐stage multi‐innovation parameter estimation for an exponential autoregressive time‐series model with moving average noise by using the data filtering technique

Gradient-based Parameter Estimation for a Nonlinear Exponential Autoregressive Time-series Model by Using the Multi-innovation

Auxiliary model maximum likelihood least squares-based iterative algorithm for multivariable autoregressive output-error autoregressive moving average systems

Multi-innovation stochastic gradient identification method

Two‐stage auxiliary model gradient‐based iterative algorithm for the input nonlinear controlled autoregressive system with variable‐gain nonlinearity

Parameter identification algorithm for ship manoeuvrability and wave peak model based multi-innovation stochastic gradient algorithm use data filtering technique

Identification of multiple-input and single-output Hammerstein controlled autoregressive moving average system based on chaotic dynamic disturbance sand cat swarm optimization