Echo State Network-Based Backstepping Adaptive Iterative Learning Control for Strict-Feedback Systems: An Error-Tracking Approach

Chen, Qiang; Shi, Huihui; Sun, Mingxuan

doi:10.1109/tcyb.2019.2931877

Cited by 114 publications

(71 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…□ Theorem 1. Considering the system (1) with the sliding mode surface (11), the control law (13), and the adaptive update laws (14) and (15), all signals of the closed-loop system are uniformly ultimately bounded within a fixed time.…”

Section: Stability Analysismentioning

confidence: 99%

“…Control. In M1, the sliding variable is designed as (11), the control law is addressed by (13), and the adaptive update laws are depicted in (14) and (15), respectively, and the parameters are listed in Table 1, where F 0 , D 0 are the initial values of F, D.…”

Section: Adaptive Fixed-time Sliding Modementioning

confidence: 99%

“…As a novel application of the small unmanned aerial vehicle, multirotor is a trending topic due to its wide advantages in practical systems [1][2][3][4][5][6][7]. So far, many research studies have been carried out on the control of multirotor systems such as robust adaptive control [8][9][10][11][12], backstepping control [13], and sliding mode control [14,15]. However, only the asymptotic stability is achieved in the aforementioned control methods.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Adaptive Fixed-Time Sliding Mode Control for Uncertain Twin-Rotor System with Experimental Validation

et al. 2019

Self Cite

View full text Add to dashboard Cite

This paper proposes an adaptive fixed-time control scheme for twin-rotor systems subject to the inertia uncertainties and external disturbances. First of all, a fixed-time sliding mode surface is constructed and the corresponding controller is developed such that the fixed-time uniform ultimate boundedness of the sliding variable and tracking error could be guaranteed simultaneously, and the setting time is independent of the initial values. The adaptive update laws are developed to estimate the upper bounds of the lumped uncertainties and external disturbances such that no prior knowledge on the system uncertainties and disturbances is required. Finally, a twin-rotor platform is constructed to verify the effectiveness of proposed scheme. Comparative results show better position tracking performance of the proposed control scheme.

show abstract

Section: Stability Analysismentioning

confidence: 99%

Section: Adaptive Fixed-time Sliding Modementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Fixed-Time Sliding Mode Control for Uncertain Twin-Rotor System with Experimental Validation

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Model‐free controllers, like the proportional‐integral‐derivative (PID) control, the sliding mode control (SMC), neural control, among others, do not require dynamic knowledge of the system. However, parameter tuning and some prior knowledge of the disturbances prevent these model‐free controllers to perform optimally.…”

Section: Introductionmentioning

confidence: 99%

Robust control under worst‐case uncertainty for unknown nonlinear systems using modified reinforcement learning

Perrusquía

2020

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

Reinforcement learning (RL) is an effective method for the design of robust controllers of unknown nonlinear systems. Normal RLs for robust control, such as actor-critic (AC) algorithms, depend on the estimation accuracy. Uncertainty in the worst case requires a large state-action space, this causes overestimation and computational problems. In this article, the RL method is modified with the k-nearest neighbor and the double Q-learning algorithm. The modified RL does not need the neural estimator as AC and can stabilize the unknown nonlinear system under the worst-case uncertainty. The convergence property of the proposed RL method is analyzed. The simulations and the experimental results show that our modified RLs are much more robust compared with the classic controllers, such as the proportional-integral-derivative, the sliding mode, and the optimal linear quadratic regulator controllers. K E Y W O R D Sk-nearest neighbors, double estimator, overestimation, robust reward, state-action space, worst-case uncertainty INTRODUCTIONThe objective of robust control is to achieve robust performance in presence of disturbances. Most robust controllers are inspired by the optimal control theory, such as  2 control, 1,2 which minimizes a certain cost function to find an optimal controller. The most popular  2 controller is the linear quadratic regulator (LQR). 3 It does not work well in presence of disturbances. The  ∞ control can find a robust controller when the system has disturbances. Its performance is poor compared with the  2 control. 4 The combination of  2 and  ∞ , called  2 ∕ ∞ control, has both advantages, that is., it has optimal performance with bounded disturbances. 5 The  2 ∕ ∞ controller design needs a complete knowledge of the system dynamics. 6 These controllers are model-based. The time-varying quadratic optimization can be calculated by the zeroing neural network. It can simultaneously achieve the finite-time convergence and inherent noise tolerance. 7 However, it requires prefect activation functions. Model-free controllers, like the proportional-integral-derivative (PID) control, 8,9 the sliding mode control (SMC), 10 neural control, 11-16 among others, do not require dynamic knowledge of the system. However, parameter tuning and some prior knowledge of the disturbances prevent these model-free controllers to perform optimally.Reinforcement learning (RL) 17,18 is another effective method without models. It is designed in the sense of  2 control. 19 The temporal difference (TD) rule, such as Q-learning, is applied to find an optimal solution for Markov decision processes. 20 The advantage of RL over the other model-free methods is that it can reach optimal performances. Recent results show that RL methods can learn  2 and  ∞ controllers without system dynamics. 21 2920The main objective of RL for  2 and  ∞ is to minimize the total accumulative reward. For a robust controller, the reward is designed in the sense of control problems  2 and  ∞ . This robust reward can be static or dynam...

show abstract

“…Recently, neural networks control has increasingly attracted attention and intensive research has been performed in adaptive law for training neural networks weights and application in different fields [1][2][3]. Neural network technique is a typical data-driven modelling method [4][5][6], which used measured data to find proper control in reversion of some expected closed-loop performance [7][8][9].…”

Section: Introductionmentioning

confidence: 99%

U-Model Based Adaptive Neural Networks Fixed-Time Backstepping Control for Uncertain Nonlinear System

Zhang

Fei

et al. 2020

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Under U-model control design framework, a fixed-time neural networks adaptive backstepping control is proposed. The majority of the previously described adaptive neural controllers were based on uniformly ultimately bounded (UUB) or practical finite stable (PFS) theory. For neural networks control, it makes the control law as well as stability analysis highly lengthy and complicated because of the unknown ideal weight and unknown approximation error. Moreover, there has been very limited research focus on adaptive law for neural networks adaptive control in finite time. Based on fixed-time stability theory, a fixed-time bounded theory is proposed for fixed-time neural networks adaptive backstepping control. The most outstanding novelty is that fixed-time adaptive law for training weights of neural networks is proposed for fixed-time neural networks adaptive control. Furthermore, by combining fixed-time adaptive law and Lyapunov-based arguments, a valid fixed-time controller design algorithm is presented with universal approximation property of neural networks to ensure the system is fixed-time bounded, rather than PFS or UUB. The controller guarantees closed-loop system fixed-time bounded in the Lyapunov sense. The benchmark simulation demonstrated effectiveness and efficiency of the proposed approach.

show abstract

Echo State Network-Based Backstepping Adaptive Iterative Learning Control for Strict-Feedback Systems: An Error-Tracking Approach

Cited by 114 publications

References 52 publications

Adaptive Fixed-Time Sliding Mode Control for Uncertain Twin-Rotor System with Experimental Validation

Adaptive Fixed-Time Sliding Mode Control for Uncertain Twin-Rotor System with Experimental Validation

Robust control under worst‐case uncertainty for unknown nonlinear systems using modified reinforcement learning

U-Model Based Adaptive Neural Networks Fixed-Time Backstepping Control for Uncertain Nonlinear System

Contact Info

Product

Resources

About