A systematic classification of neural-network-based control

Agarwal, Mukul

doi:10.1109/37.581297

Cited by 114 publications

(2 citation statements)

References 147 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many studies have been conducted on the application of neural networks (NNs) to predict, recognize, and control dynamic systems [36]- [40]. One of the most outstanding advantages of NN is its ability to approximate arbitrary linear or nonlinear systems through learning.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Single-Input Recurrent WCMAC-Based Supervisory Control for De-icing Robot Manipulator

Ngo,

Le,

Lam

et al. 2023

Journal of Robotics and Control

View full text Add to dashboard Cite

The control of any robotic system always faces many great challenges in theory and practice. Because between theory and reality, there is always a huge difference in the uncertainty components in the system. That leads to the accuracy and stability of the system not being guaranteed with the set requirements. This paper presents a novel adaptive single-input recurrent wavelet differentiable cerebellar model articulation controller (S-RWCMAC)-based supervisory control system for an m-link robot manipulator to achieve precision trajectory tracking. This adaptive S-RWCMAC-based supervisory control system consists of a main adaptive S-RWCMAC, a supervisory controller, and an adaptive robust controller. The S-RWCMAC incorporates the advantages of the wavelet decomposition property with a CMAC fast learning ability, dynamic response, and input space dimension of RWCMAC can be simplified; and it is used to control the plant. The supervisory controller is appended to the adaptive S-RWCMAC to force the system states within a predefined constraint set and the adaptive robust controller is developed to dispel the effect of the approximate error. In this scheme, if the adaptive S-RWCMAC can not maintain the system states within the constraint set. Then, the supervisory controller will work to pull the states back to the constraint set and otherwise is idle. The online tuning laws of S-RWCMAC and the robust controller parameters are derived from the gradient-descent learning method and Lyapunov function so that the stability of the system can be guaranteed. The simulation and experimental results of the novel three-link De-icing robot manipulator are provided to verify the effectiveness of the proposed control methodology. The results indicate that the proposed model has superior accuracy compared to that of the Standalone CMAC Controller. The parameters of the average squared error in the S-RWCMAC -based 3 robot joints are lower than those of the Standalone CMAC Controller by 0.023%, 0.029%, and 0.032%, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Adaptive Single-Input Recurrent WCMAC-Based Supervisory Control for De-icing Robot Manipulator

Ngo,

Le,

Lam

et al. 2023

Journal of Robotics and Control

View full text Add to dashboard Cite

show abstract

“…Recently, Artificial Neural Networks (ANNs) are utilized as an alternative and attractive tool for steady-state/dynamic nonlinear process modeling and model-based control in situations where the development of phenomenological or empirical regression models becomes impractical or cumbersome (Agarwal, 1997;Şahin & Öztürk, 2018;Taheri-Garavand, Meda, & Naderloo, 2018). ANN is reported as an "effortless computational" intelligent system inspired by the biological neural structure of the brain.…”

Section: Introductionmentioning

confidence: 99%

Modeling and optimization of cooking process parameters to improve the nutritional profile of fried fish by robust hybrid artificial intelligence approach

Sadhu

Banerjee

Lahiri

et al. 2020

J Food Process Engineering

View full text Add to dashboard Cite

Fish, being a good source of nutrients, is often cooked by different methods before consumption, which affect the beneficial quality detrimentally. In this study, Catla catla, and mustard oil are selected as representative of fish and cooking oil for frying, respectively, because of their agricultural importance and worldwide demand. Extensive experiments are performed varying the effective processing variables of conventional frying viz., temperature (140 °C‐240 °C), time (5 min–20 min) and oil amount (25 ml/kg of fish‐100 ml/kg of fish) to correlate the drastic reduction of the nutritional quality indices, that is, ω‐3/ω‐6 and cis/trans‐fatty acids (FAs) profiles of fish after frying. To establish a nonlinear correlation between these inputs and outputs, an exhaustive search of all available artificial neural network (ANN) algorithms and activation functions is executed for the development of a model. The hybrid robust process approach integrating ANN with differential evolution (DE) and simulated annealing (SA) are employed to optimize the cooking parameters for regaining nutritional impact. After frying ω‐3/ω‐6 and cis/trans‐FAs ratio deteriorated by 76.65% and 92.68%, respectively, than the fresh samples. The ANN‐DE and ANN‐SA formalism efficiently enhanced these nutritional parameters up to 33.18% and 79%, respectively. Practical applications The present study applied artificial neural network (ANN) as an advanced alternative modeling tool to propose a generalized nonlinear correlation between temperature, time, oil amount, and nutritional values, that is, ω‐3/ω‐6 and cis/trans‐fatty acids (FAs) profiles of fried fish. Frying time provided a strong impact on food nutrition compared to other two input variables. Frying process detrimentally affected both the nutritional indices, that is, ω‐3/ω‐6 and cis/trans‐FAs profiles. The meta‐heuristic, stochastic optimization algorithms, namely differential evolution and simulated annealing along with ANN‐based processed model were implemented successfully to tune the cooking parameters, so that food quality indices of fish improved again to maximum value. The artificial intelligence modeling, along with optimizing methodology based parameters tuning approach described here is generic and can be advantageously extended to other experimentation of food process engineering. Besides, the finding of this study will benefit common people also.

show abstract

Neural adaptive control of non-linear plants via a multiple inverse model approach

Zufiria

Fraile-Ardanuy

Riaza

et al. 1999

Int. J. Adapt. Control Signal Process.

View full text Add to dashboard Cite

SUMMARYIn this paper, neural architectures for controlling non-linear plants with parameter variation are proposed. In the "rst part of the document, the concept of specialized learning over an operation region is considered in order to identify the inverse dynamics of a given plant. Some aspects concerning discretization and invertibility of continuous-time plants are also addressed. In the second part of this work, a control architecture which combines the former approach of inverse identi"cation through specialised learning with a multiple model scheme is presented. Finally, simulation results are discussed, evaluating the performance of the proposed schemes; speci"cally, the presented controllers are applied to the simulation of the control of a robot arm.

show abstract

A systematic classification of neural-network-based control

Cited by 114 publications

References 147 publications

Adaptive Single-Input Recurrent WCMAC-Based Supervisory Control for De-icing Robot Manipulator

Adaptive Single-Input Recurrent WCMAC-Based Supervisory Control for De-icing Robot Manipulator

Modeling and optimization of cooking process parameters to improve the nutritional profile of fried fish by robust hybrid artificial intelligence approach

Neural adaptive control of non-linear plants via a multiple inverse model approach

Contact Info

Product

Resources

About