Three-level hierarchical model-free learning approach to trajectory tracking control

Rădac, Mircea-Bogdan; Precup, Radu‐Emil

doi:10.1016/j.engappai.2016.06.009

Cited by 35 publications

(9 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…JITL method has been largely applied in nonlinear processes modeling and control, [29][30][31][32][33][34] which is built on the basis of database and local modeling techniques. In JITL algorithm, there are three main steps implemented to predict output.…”

Section: Jitl Algorithmmentioning

confidence: 99%

A local dynamic extreme learning machine based iterative learning control of nonlinear batch process

Zhou

Jia

2021

Optim Control Appl Methods

View full text Add to dashboard Cite

This article deals with the optimal control issue of nonlinear batch process. First, in order to derive high efficiency and accuracy process model, a novel hierarchical searching mechanism local dynamic nonlinear model is constructed which is composed of just‐in‐time learning and extreme learning machine (JITL‐ELM). Then, based on the local dynamic JITL‐ELM model, an optimal quadratic‐criterion‐based iterative learning control (Q‐ILC) algorithm is presented, where the control input trajectory can be obtained by solving a quadratic programming problem. Moreover, on the basis of inverse model system, the initial batch control input trajectory of the Q‐ILC algorithm can be obtained by the use of JITL method. As a result, not only the issue of model‐plant mismatch and real‐time disturbance can be solved, but also obtain faster system convergence rate and smaller tracking error. Besides, the convergence properties of control input and tracking error are analyzed. Finally, a typical batch process is presented to demonstrate the feasibility and superiority.

show abstract

Section: Jitl Algorithmmentioning

confidence: 99%

A local dynamic extreme learning machine based iterative learning control of nonlinear batch process

Zhou

Jia

2021

Optim Control Appl Methods

View full text Add to dashboard Cite

show abstract

“…Using this fact, it follows from (10) that the ORM tracking errors s( An NN can be used as a controller for nonlinear state-feedback control learning. Nonlinear VRFT is proposed in [41,42] and successfully applied to NN controllers in [41,[43][44][45] but only for output feedback control and not for state-feedback control as in here.…”

Section: Nonlinear State-feedback Vrft For Approximate Orm Tracking Cmentioning

confidence: 99%

Data-Driven Model-Free Tracking Reinforcement Learning Control with VRFT-based Adaptive Actor-Critic

Rădac

Precup

2019

Applied Sciences

View full text Add to dashboard Cite

This paper proposes a neural network (NN)-based control scheme in an Adaptive Actor-Critic (AAC) learning framework designed for output reference model tracking, as a representative deep-learning application. The control learning scheme is model-free with respect to the process model. AAC designs usually require an initial controller to start the learning process; however, systematic guidelines for choosing the initial controller are not offered in the literature, especially in a model-free manner. Virtual Reference Feedback Tuning (VRFT) is proposed for obtaining an initially stabilizing NN nonlinear state-feedback controller, designed from input-state-output data collected from the process in open-loop setting. The solution offers systematic design guidelines for initial controller design. The resulting suboptimal state-feedback controller is next improved under the AAC learning framework by online adaptation of a critic NN and a controller NN. The mixed VRFT-AAC approach is validated on a multi-input multi-output nonlinear constrained coupled vertical two-tank system. Discussions on the control system behavior are offered together with comparisons with similar approaches.

show abstract

“…knowledge-based systems and statistical inference, which seem to be the most widespread). The authors of the present paper have been interested in applying them to medical systems [3], [4] and modeling with focus on medical [5], [6] and non-medical process control [7]- [14] including novel training algorithms formulated in the framework of Iterative Learning Control (ILC) [7]- [11] and model predictive control [12]. Their conclusions have been similar to those emphasized by the main researchers in AI domain [1], [2]: these conventional mechanisms have remarkable advantages, but they are suitable only for a small area of medical decision-making problems.…”

Section: Introductionmentioning

confidence: 99%

“…, and also the data sample index, and it is also dropped out as follows if the simplification of notations is intended, y is the output, W is the vector of output layer weights,  is the vector of outputs of hidden layer neurons: (11) and the first term in  given in (10) corresponds to the bias of the output neuron. Each neuron in the hidden layer neuron is parameterized by the vector of weights V m , m = 1...H, and…”

mentioning

confidence: 99%

Results and Challenges of Artificial Neural Networks Used for Decision-Making and Control in Medical Applications

Albu¹,

Precup

Teban³

2019

FU Mech Eng

Self Cite

View full text Add to dashboard Cite

The aim of this paper is to present several approaches by which technology can assist medical decision-making. This is an essential, but also a difficult activity, which implies a large number of medical and technical aspects. But, more important, it involves humans: on the one hand, the patient, who has a medical problem and who requires the best solution; on the other hand, the physician, who should be able to provide, in any circumstances, a decision or a prediction regarding the current and the future medical status of the patient. The technology, in general, and particularly the Artificial Intelligence (AI) tools could help both of them, and it is assisted by appropriate theory regarding modeling tools. One of the most powerful mechanisms that can be used in this field is the Artificial Neural Networks (ANNs). This paper presents some of the results obtained by the Process Control group of the Politehnica University Timisoara, Romania, in the field of ANNs applied to modeling, prediction and decision-making related to medical systems. An Iterative Learning Control-based approach to batch training a feedforward ANN architecture is given. The paper includes authors’ concerns in this domain and emphasizes that these intelligent models, even if they are artificial, are able to make decisions, being useful tools for prevention, early detection and personalized healthcare.

show abstract

Three-level hierarchical model-free learning approach to trajectory tracking control

Cited by 35 publications

References 51 publications

A local dynamic extreme learning machine based iterative learning control of nonlinear batch process

A local dynamic extreme learning machine based iterative learning control of nonlinear batch process

Data-Driven Model-Free Tracking Reinforcement Learning Control with VRFT-based Adaptive Actor-Critic

Results and Challenges of Artificial Neural Networks Used for Decision-Making and Control in Medical Applications

Contact Info

Product

Resources

About