Bio-inspired spiking neural network for nonlinear systems control

Fernández, Javier Pérez; Carrillo, Juan Antonio Cabrera; Castillo, Juan José; Velasco, Juan M.

doi:10.1016/j.neunet.2018.04.002

Cited by 25 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wang et al [54] described an approach to designing a PID controller for a system with delay using a new NN-based Smith predictor, which makes the controller adaptive to changing system parameters. In [55,56], spike NNs are used to develop a controller for a highly nonlinear control system, which uses time-based pulse trains and can perform faster and more complex calculations. Spike NN controllers are capable of online learning and self-adaptation when moving from simulation to the real world.…”

Section: Stabilization and Program Control Problemsmentioning

confidence: 99%

Methods of Intelligent Control in Mechatronics and Robotic Engineering: A Survey

Zaitceva

Andrievsky

2022

Electronics

View full text Add to dashboard Cite

Artificial intelligence is becoming an increasingly popular tool in more and more areas of technology. New challenges in control systems design and application are related to increased productivity, control flexibility, and processing of big data. Some kinds of systems require autonomy in real-time decision-making, while the other ones may serve as an essential factor in human-robot interaction and human influences on system performance. Naturally, the complex tasks of controlling technical systems require new modern solutions, but there remains an inextricable link between control theory and artificial intelligence. The first part of the present survey is devoted to the main intelligent control methods in technical systems. Among them, modern methods of adaptive and optimal control, fuzzy logic, and machine learning are considered. In its second part, the crucial achievements in intelligent control applications in robotic and mechatronic systems over the past decade are considered. The references are structured according to the type of such common control problems as stabilization, controller tuning, identification, parametric optimization, iterative learning, and prediction. In the conclusion, the main problems and tendencies toward intelligent control methods improvement are outlined.

show abstract

Section: Stabilization and Program Control Problemsmentioning

confidence: 99%

Methods of Intelligent Control in Mechatronics and Robotic Engineering: A Survey

Zaitceva

Andrievsky

2022

Electronics

View full text Add to dashboard Cite

show abstract

“…In particular, the differential evolution (DE) metaheuristic algorithm is used to perform the design of Q NN . This choice is justified by the fact that DE has proven to be effective in the training of neural networks [20,21] and that it has shown an outstanding performance in the design of dynamic quantizers [9]. DE is a population-based metaheuristic algorithm inspired in the mechanism of biological evolution [22,23].…”

Section: A T E X I T S H a 1 _ B A S E 6 4 = "mentioning

confidence: 99%

Design of Neural Network Quantizers for Networked Control Systems

Ramirez

Minami

2019

Electronics

View full text Add to dashboard Cite

Nowadays, networked control systems (NCSs) are being widely implemented in many applications. However, several problems negatively affect and compromise the design of practical NCSs. One of them is the performance degradation of the system due to quantization. This paper aims to develop dynamic quantizers for NCSs and their design methods that alleviate the effects of the quantization problem. In this paper, we propose a type of dynamic quantizers implemented with neural networks and memories, which can be tuned by a time series data of the plant inputs and outputs. Since the proposed quantizer can be designed without the model information of the system, the quantizer could be applied to any system with uncertainty or nonlinearity. This paper gives two types of quantizers, and they differ from each other in the neural networks structure. The effectiveness of these quantizers and their design method are verified using numerical examples. Besides, their performances are compared among each other using statistical analysis tools.

show abstract

“…For example, in spiking NNs a temporal spike is used for mapping highly nonlinear dynamic models. Networks of spiking neurons are, with regard to the number of neurons, computationally more powerful than earlier ANN models [3]. Nevertheless, training such networks is difficult due to the non-differentiable nature of spike events [4].…”

Section: Introductionmentioning

confidence: 99%

“…Another relevant class of ANN which exhibits temporal dynamics is that of recurrent (cyclic) NNs (RNNs). Unlike feedforward (acyclic) NNs, recurrent NNs can use their internal state (memory) to process sequences of inputs, creating and processing memories of arbitrary sequences of input patterns [3]. A special class of recurrent NN is based on the Long Short-Term Memory (LSTM) unit, which is made of a cell, an input gate, an output gate and a forget gate.…”

Section: Introductionmentioning

confidence: 99%

“…A special class of recurrent NN is based on the Long Short-Term Memory (LSTM) unit, which is made of a cell, an input gate, an output gate and a forget gate. The cell remembers values over arbitrary time intervals and the three gates regulate the flow of information into and out of the cell [3]. However, the training process of recurrent NNs strongly depends on the set of constraints and regularizations used in the optimization process, resulting in a not entirely unbiased task with respect to the "how" [5].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Using Stigmergy to Incorporate the Time into Artificial Neural Networks

Galatolo

Cimino

Vaglini

2018

Mining Intelligence and Knowledge Exploration

View full text Add to dashboard Cite

A current research trend in neurocomputing involves the design of novel artificial neural networks incorporating the concept of time into their operating model. In this paper, a novel architecture that employs stigmergy is proposed. Computational stigmergy is used to dynamically increase (or decrease) the strength of a connection, or the activation level, of an artificial neuron when stimulated (or released). This study lays down a basic framework for the derivation of a stigmergic NN with a related training algorithm. To show its potential, some pilot experiments have been reported. The XOR problem is solved by using only one single stigmergic neuron with one input and one output. A static NN, a stigmergic NN, a recurrent NN and a long short-term memory NN have been trained to solve the MNIST digits recognition benchmark.

show abstract

Bio-inspired spiking neural network for nonlinear systems control

Cited by 25 publications

References 25 publications

Methods of Intelligent Control in Mechatronics and Robotic Engineering: A Survey

Methods of Intelligent Control in Mechatronics and Robotic Engineering: A Survey

Design of Neural Network Quantizers for Networked Control Systems

Using Stigmergy to Incorporate the Time into Artificial Neural Networks

Contact Info

Product

Resources

About