Neural network–based direct adaptive robust control of unknown MIMO nonlinear systems using state observer

Cheng, Cheng; Liu, Songyong; Wu, Hongzhuang; Zhang, Ying

doi:10.1002/acs.3064

Cited by 12 publications

(22 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Example Consider the following single‐link robot dynamic equations

\begin{align} M \overset{q}{true} + \frac{1}{2} m g L normalsin q = u \\ y = q, \end{align}

where M =0.5 is the moment of inertia, m =1 and L =1 denote the link mass and the length respectively, g =9.8 m/s is the gravity coefficient , q is the angle of the link, and u is the control torque.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…Remark Compared with the design in Reference where the tracking error z 1 is defined as

z_{1} = {\hat{x}}_{1} - y_{d}

, the definition of z 1 = y − y d is more reasonable and can achieve better tracking performance easily. In addition, the prescribed performance will be guaranteed by introducing an NPPF and an ETF in this work, which is difficult to achieve with the definition

z_{1} = {\hat{x}}_{1} - y_{d}

in Reference , and the filtering errors

ζ_{i} = {\overline{α}}_{i} - α_{i}

are not considered in the stability analysis in Reference .…”

Section: Adaptive Output Feedback Control Designmentioning

confidence: 99%

“…In Referecne , an adaptive fuzzy output feedback tracking back‐stepping control scheme was proposed for a class of strict‐feedback nonlinear systems with immeasurable states and unknown dead zone. In Reference , a coupled design approach between NDO and state observer was studied for a class of uncertain SISO nonlinear systems, and an adaptive neural output feedback back‐stepping control scheme was proposed based on the output of NDO and NN. For a class of uncertain MIMO nonlinear system, a novel observer‐based direct adaptive neuro‐sliding mode control approach was presented in Reference .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Adaptive neural prescribed performance output feedback control of pure feedback nonlinear systems using disturbance observer

Chen

Yang

2020

Adaptive Control & Signal

View full text Add to dashboard Cite

Summary In this study, an adaptive output feedback control with prescribed performance is proposed for unknown pure feedback nonlinear systems with external disturbances and unmeasured states. A novel prescribed performance function is developed and incorporated into an output error transformation to achieve tracking control with prescribed performance. To handle the unknown non‐affine nonlinearities and avoid the algebraic loop problem, the radial basis function neural network (RBFNN) is adopted to approximate the unknown non‐affine nonlinearities with the help of Butterworth low‐pass filter. Based on the output of the RBFNN, the coupled design between sate observer and disturbance observer is presented to estimate the unmeasured states and compounded disturbances. Then, the adaptive output feedback control scheme is proposed for unknown pure feedback nonlinear systems, where a first‐order filter is introduced to tackle with the issue of “explosion of complexity” in the traditional back‐stepping approach. The boundedness and convergence of the closed‐loop system are proved rigorously by utilizing the Lyapunov stability theorem. Finally, simulation studies are worked out to demonstrate the effectiveness of the proposed scheme.

show abstract

“…Example Consider the following single‐link robot dynamic equations

\begin{align} M \overset{q}{true} + \frac{1}{2} m g L normalsin q = u \\ y = q, \end{align}

Section: Simulation Resultsmentioning

confidence: 99%

“…Remark Compared with the design in Reference where the tracking error z 1 is defined as

z_{1} = {\hat{x}}_{1} - y_{d}

z_{1} = {\hat{x}}_{1} - y_{d}

in Reference , and the filtering errors

ζ_{i} = {\overline{α}}_{i} - α_{i}

are not considered in the stability analysis in Reference .…”

Section: Adaptive Output Feedback Control Designmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive neural prescribed performance output feedback control of pure feedback nonlinear systems using disturbance observer

Chen

Yang

2020

Adaptive Control & Signal

View full text Add to dashboard Cite

show abstract

“…In recent years, many scholars applies intelligent control technology to engineering practice. In literature [31], an adaptive linear neuron neural network is introduced into the traditional sliding mode observer, which has good performance. In order to reduce chattering in real time, an adaptive fuzzy logic is designed to Approach the parameters of the sliding mode controller [32].…”

Section: Introductionmentioning

confidence: 99%

Dynamics Modeling and Tension Control of Composites Winding System Based on ASMC

2020

View full text Add to dashboard Cite

Tension is a key processing parameter in the process of composite fiber tape winding. The fluctuation of tension will affect the accuracy of winding and the performance of winding products, such as stress uniformity, fatigue resistance, strength, compactness and resin content. In view of the winding tension is time-varying and the application of tension needs to be more accurate, a tension calculation model is established. Due to the influence of dynamic performance of tension control system by cogging torque, inverter dead zone, ring gear clearance, friction torque, parameter drift, and measurement noise, an adaptive sliding mode control (ASMC) based on extended state observer (ESO) is proposed. The stability of the closed-loop system is proven by Lyapunov theory, the system state variables is estimated by ESO, and the input dead-zone is compensated by the designed adaptive law. Simulation and experimental results show that ESO-based ASMC improves the robustness and dynamic response performance of the tension control system, and can effectively suppress the chattering of the sliding mode control system. The void content and residual stress of composite products have been improved obviously. INDEX TERMS Composites tape winding, dynamics modeling, ESO, ASMC, tension control.

show abstract

“…[24][25][26] In addition, it is worth noting that the neural network has strong nonlinear function approximation ability and has been widely used in control system design. [27][28][29][30][31] Radial basis function (RBF) neural network was always used to approximate the nonlinear functions in robot system dynamics brought by the unknown environment. 30,32 Therefore, we choose to use RBF neural network to approximate coupling terms for uncertainties and faults.…”

Section: Introductionmentioning

confidence: 99%

Modeling and distributed adaptive fault‐tolerant vibration control for bridge beam with single‐parameter adaptive neural network

Gao

Yang

Liu

2020

Adaptive Control & Signal

View full text Add to dashboard Cite

Modeling and vibration control of a bridge beam system are considered in this article. The beam bridge with both ends fixed can be regarded as an Euler-Bernoulli beam, which is a typical distributed parameter system. First, the partial differential equations (PDE) model of the bridge was established according to the Hamilton principle. Then, a reasonable distributed control law was designed on the PDE model to eliminate the elastic deformation and suppress the vibration of the bridge. At the same time, uncertainties related to system status were considered during the design of the closed-loop system. In addition, the possible actuator and sensor faults in the control system were analyzed. Single-parameter adaptive neural networks were used to estimate the effects of coupling terms for uncertainties and faults. The parameter estimation adaptive law was designed to replace the adjustment of neural network weights, which simplifies the algorithm and facilitates practical engineering applications. Finally, the feasibility of the control system was verified by simulation.

show abstract

Neural network–based direct adaptive robust control of unknown MIMO nonlinear systems using state observer

Cited by 12 publications

References 40 publications

Adaptive neural prescribed performance output feedback control of pure feedback nonlinear systems using disturbance observer

Adaptive neural prescribed performance output feedback control of pure feedback nonlinear systems using disturbance observer

Dynamics Modeling and Tension Control of Composites Winding System Based on ASMC

Modeling and distributed adaptive fault‐tolerant vibration control for bridge beam with single‐parameter adaptive neural network

Contact Info

Product

Resources

About