To see the other types of publications on this topic, follow the link: Neural network controller.
Journal articles on the topic 'Neural network controller'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Neural network controller.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Leal, Hugo M., Ramiro S. Barbosa, and Isabel S. Jesus. "Control of a Mobile Line-Following Robot Using Neural Networks." Algorithms 18, no. 1 (2025): 51. https://doi.org/10.3390/a18010051.
Full textAbstract:
This work aims to develop and compare the performance of a line-following robot using both neural networks and classical controllers such as Proportional–Integral–Derivative (PID). Initially, the robot’s infrared sensors were employed to follow a line using a PID controller. The data from this method were then used to train a Long Short-Term Memory (LSTM) network, which successfully replicated the behavior of the PID controller. In a subsequent experiment, the robot’s camera was used for line-following with neural networks. Images of the track were captured, categorized, and used to train a convolutional neural network (CNN), which then controlled the robot in real time. The results showed that neural networks are effective but require more processing and calibration. On the other hand, PID controllers proved to be simpler and more efficient for the tested tracks. Although neural networks are very promising for advanced applications, they are also capable of handling simpler tasks effectively.
2
Gücüyener, İsmet. "Fuzzy Neural-Network-Based Controller." Solid State Phenomena 220-221 (January 2015): 407–12. http://dx.doi.org/10.4028/www.scientific.net/ssp.220-221.407.
Full textAbstract:
Using a controller is necessary for any automation system. The controller must be cheap, reliable, user friendly and not cause any problems for inputs and outputs. Classical control systems like proportional integral derivative (PID) put adequate results of linear systems and continuous-time. In fact, real control systems are time-variant, with non-linearity and poorly calculated dynamic variables. For this reason, conventional control systems need an expert person to adjust controller parameters in general. Sometimes an operator is required to solve control problems. Human control is not completely reliable. Also, it does not include any electronic communication. In modern factories, every point must be monitored and electronically controlled from remote points when necessary. In this study, including every electronic communication channel, a simplified handling, low-cost, reliable, Fuzzy Neural Network Controller (FNNC) is designed.
3
Yamada, Takayuki, and Tetsuro Yabuta. "Adaptive Neural Network Controllers for Dynamics Systems." Journal of Robotics and Mechatronics 2, no. 4 (1990): 245–57. http://dx.doi.org/10.20965/jrm.1990.p0245.
Full textAbstract:
Many studies such as Kawato's work have been undertaken in order to apply both the flexibility and learning ability of neural networks to dynamic system controllers. However, their characteristics have not yet been completely clarified. On the other hand, many studies have established conventional control theories such as adaptive control. If we can clarify the relationship between neural network controllers and adaptive controllers, the two control algorithms will be developed considerably by making use of the advantages of each. Therefore, this paper proposes a neural network direct controller in order to construct an interface between neural network and conventional control theories. This paper also proposes an open loop type of controller in order to realize inverse dynamics using only the neural network. Analytical approaches prove the local stability of the proposed controllers. Simulated and experimental results verify their realization and confirm their characteristics. This paper also discusses the relationship between neural network controllers and adaptive controllers.
4
Zhang, Yong Chao, Wen Zhuang Zhao, and Jin Lian Chen. "The Research and Application of the Fuzzy Neural Network Control Based on Genetic Algorithm." Advanced Materials Research 403-408 (November 2011): 191–95. http://dx.doi.org/10.4028/www.scientific.net/amr.403-408.191.
Full textAbstract:
How fuzzy technology and neural networks and genetic algorithm combine with each other has become the focus of research. A fuzzy neural network controller was proposed based on defuzzification and optimization around the fuzzy neural network structure. Genetic algorithm of fuzzy neural network was brought forward based on optimal control theory. Optimal structure and parameters of fuzzy neural network controller were Offline searched by way of controller performance indicators of genetic algorithm. Fuzzy neural network controller through genetic algorithm was accessed in fuzzy neural network intelligent control system.
5
Sunar, M., A. M. A. Gurain, and M. Mohandes. "Substructural neural network controller." Computers & Structures 78, no. 4 (2000): 575–81. http://dx.doi.org/10.1016/s0045-7949(00)00039-0.
Full text
6
Ruano, A. E. B., D. I. Jones, and P. J. Fleming. "A neural network controller." IFAC Proceedings Volumes 24, no. 7 (1991): 27–32. http://dx.doi.org/10.1016/b978-0-08-041699-1.50009-4.
Full text
7
Karunasena, G. M. K. B., H. D. N. S. Priyankara, and B. G. D. A. Madhusank. "Artificial Neural Network vs PID Controller for Magnetic Levitation System." International Journal of Innovative Science and Research Technology 5, no. 7 (2020): 505–11. http://dx.doi.org/10.38124/ijisrt20jul432.
Full textAbstract:
This research investigates the acceptability of the Artificial Neural Networks (ANN) over the PID Controller for the control of the Magnetic Levitation System (MLS). In the field of advanced control systems, this system identifies as a feedback-controlled, single input- single output (SISO) system. This SISO system used a PID controller for vertical trajectory controlling of a metal sphere in airspace by using an electromagnetic force that directed to upward. The vertical position of the metal sphere controlled according to the applied magnetic force generated by a powerful electromagnet and the electromagnetic force controlled by varying the supply voltage. To control this nonlinear system, we develop a multilayer artificial neural network by using Matlab software and integrate that with the physical magnetic levitation model. According to specific initial conditions, the actual responses of the magnetic levitation system with artificial neural network compares the desire response of the metal sphere. The ability of control this nonlinear system by using neural networks validate by comparing results obtained by the PID controller and artificial neural network.
8
Lippe, Wolfram-M., Steffen Niendieck, and Andreas Tenhagen. "On the Optimization of Fuzzy-Controllers by Neural Networks." Journal of Advanced Computational Intelligence and Intelligent Informatics 3, no. 3 (1999): 158–63. http://dx.doi.org/10.20965/jaciii.1999.p0158.
Full textAbstract:
Methods are known for combining fuzzy-controllers with neural networks. One of the reasons of these combinations is to work around the fuzzy controllers’ disadvantage of not being adaptive. It is helpful to represent a given fuzzy controller by a neural network and to have rules adapted by a special learning algorithm. Some of these methods are applied in the NEFCONmode or the model of Lin and Lee. Unfortunately, none adapts all fuzzy-controller components. We suggest a new model enabling the user to represent a given fuzzy controller by a neural network and adapt its components as desired.
9
Jiang, Yiming, Chenguang Yang, Shi-lu Dai, and Beibei Ren. "Deterministic learning enhanced neutral network control of unmanned helicopter." International Journal of Advanced Robotic Systems 13, no. 6 (2016): 172988141667111. http://dx.doi.org/10.1177/1729881416671118.
Full textAbstract:
In this article, a neural network–based tracking controller is developed for an unmanned helicopter system with guaranteed global stability in the presence of uncertain system dynamics. Due to the coupling and modeling uncertainties of the helicopter systems, neutral networks approximation techniques are employed to compensate the unknown dynamics of each subsystem. In order to extend the semiglobal stability achieved by conventional neural control to global stability, a switching mechanism is also integrated into the control design, such that the resulted neural controller is always valid without any concern on either initial conditions or range of state variables. In addition, deterministic learning is applied to the neutral network learning control, such that the adaptive neutral networks are able to store the learned knowledge that could be reused to construct neutral network controller with improved control performance. Simulation studies are carried out on a helicopter model to illustrate the effectiveness of the proposed control design.
10
Liang, Haijun, Changyan Liu, Kuanming Chen, Jianguo Kong, Qicong Han, and Tiantian Zhao. "Controller Fatigue State Detection Based on ES-DFNN." Aerospace 8, no. 12 (2021): 383. http://dx.doi.org/10.3390/aerospace8120383.
Full textAbstract:
The fatiguing work of air traffic controllers inevitably threatens air traffic safety. Determining whether eyes are in an open or closed state is currently the main method for detecting fatigue in air traffic controllers. Here, an eye state recognition model based on deep-fusion neural networks is proposed for determination of the fatigue state of controllers. This method uses transfer learning strategies to pre-train deep neural networks and deep convolutional neural networks and performs network fusion at the decision-making layer. The fused network demonstrated an improved ability to classify the target domain dataset. First, a deep-cascaded neural network algorithm was used to realize face detection and eye positioning. Second, according to the eye selection mechanism, the pictures of the eyes to be tested were cropped and passed into the deep-fusion neural network to determine the eye state. Finally, the PERCLOS indicator was combined to detect the fatigue state of the controller. On the ZJU, CEW and ATCE datasets, the accuracy, F1 score and AUC values of different networks were compared, and, on the ZJU and CEW datasets, the recognition accuracy and AUC values among different methods were evaluated based on a comparative experiment. The experimental results show that the deep-fusion neural network model demonstrated better performance than the other assessed network models. When applied to the controller eye dataset, the recognition accuracy was 98.44%, and the recognition accuracy for the test video was 97.30%.
11
Burakov, M. V., V. F. Shishlakov, and A. S. Konovalov. "ADAPTIVE NEURAL NETWORK PID CONTROLLER." Issues of radio electronics, no. 10 (October 20, 2018): 86–92. http://dx.doi.org/10.21778/2218-5453-2018-10-86-92.
Full textAbstract:
The problem of constructing an adaptive PID controller based on the Hopfield neural network for a linear dynamic plant of the second order is considered. A description of the plant in the form of a discrete transfer function is used, the coefficients of which are determined with the help of a neural network that minimizes the discrepancy between the outputs of the plant and the model. The neural network processes the current and delayed input and output signals of the plant, forming an output for estimating the coefficients of the model. Another neural network determines the PID regulator coefficients at which the dynamics of the system approach the dynamics of the reference process. The calculation of weights and displacements of neurons in Hopfield networks used for identification and control is based on the construction of Lyapunov functions. The proposed methodology can be used to organize adaptive control of a wide class of linear dynamic systems with variable parameters. The results of the simulation in the article show the effectiveness of the proposed method.
12
Jung, Seul, and T. C. Hsia. "Neural network inverse control techniques for PD controlled robot manipulator." Robotica 18, no. 3 (2000): 305–14. http://dx.doi.org/10.1017/s0263574799002064.
Full textAbstract:
In this paper neural network (NN) control techniques for non-model based PD controlled robot manipulators are proposed. The main difference between the proposed technique and the existing feedback error learning (FEL) technique is that compensation of robot dynamics uncertainties is done outside the control loop by modifying the desired input trajectory. By using different NN training signals, two NN control schemes are developed. One is comparable to that in the FEL technique and another has to deal with the Jacobian of the PD controlled robot dynamic system. Performances of both controllers for various trajectories with different PD controller gains are examined and compared with that of the FEL controller. It is shown that the new control technique performed better and robust to PD controller gain variations.
13
Günther, Johannes, Elias Reichensdörfer, Patrick M. Pilarski, and Klaus Diepold. "Interpretable PID parameter tuning for control engineering using general dynamic neural networks: An extensive comparison." PLOS ONE 15, no. 12 (2020): e0243320. http://dx.doi.org/10.1371/journal.pone.0243320.
Full textAbstract:
Modern automation systems largely rely on closed loop control, wherein a controller interacts with a controlled process via actions, based on observations. These systems are increasingly complex, yet most deployed controllers are linear Proportional-Integral-Derivative (PID) controllers. PID controllers perform well on linear and near-linear systems but their simplicity is at odds with the robustness required to reliably control complex processes. Modern machine learning techniques offer a way to extend PID controllers beyond their linear control capabilities by using neural networks. However, such an extension comes at the cost of losing stability guarantees and controller interpretability. In this paper, we examine the utility of extending PID controllers with recurrent neural networks—–namely, General Dynamic Neural Networks (GDNN); we show that GDNN (neural) PID controllers perform well on a range of complex control systems and highlight how they can be a scalable and interpretable option for modern control systems. To do so, we provide an extensive study using four benchmark systems that represent the most common control engineering benchmarks. All control environments are evaluated with and without noise as well as with and without disturbances. The neural PID controller performs better than standard PID control in 15 of 16 tasks and better than model-based control in 13 of 16 tasks. As a second contribution, we address the lack of interpretability that prevents neural networks from being used in real-world control processes. We use bounded-input bounded-output stability analysis to evaluate the parameters suggested by the neural network, making them understandable for engineers. This combination of rigorous evaluation paired with better interpretability is an important step towards the acceptance of neural-network-based control approaches for real-world systems. It is furthermore an important step towards interpretable and safely applied artificial intelligence.
14
Jiang, Yiming, Chenguang Yang, Min Wang, Ning Wang, and Xiaofeng Liu. "Bioinspired control design using cerebellar model articulation controller network for omnidirectional mobile robots." Advances in Mechanical Engineering 10, no. 8 (2018): 168781401879434. http://dx.doi.org/10.1177/1687814018794349.
Full textAbstract:
As a learning mechanism that emulates the structure of the cerebellum, cerebellar model articulation controllers have been widely adopted in the control of robotic systems because of the fast learning ability and simple computational structure. In this article, a cerebellar model articulation controller–based neural network controller is developed for an omnidirectional mobile robot. With the powerful learning ability of cerebellar model articulation controller, a cerebellar model articulation controller neural network is constructed to learn the complex dynamics of the omnidirectional mobile robot such that the robot is controlled without a priori knowledge of the robot dynamics. In addition, to overcome the limitation of the neural network controller, a global control technique with a group of smooth switching functions is designed such that the global ultimately uniformly boundedness of cerebellar model articulation controller is achieved instead of conventional semi-global ultimately uniformly boundedness. Moreover, smooth decreasing boundary functions are synthesized into the controller to guarantee the transient control performance. Based on an omnidirectional mobile robot, numerical experiments have been conducted to demonstrate the effectiveness of the proposed cerebellar model articulation controller controller.
15
Shi, De Quan, Gui Li Gao, Ying Liu, Hui Ying Tang, and Zhi Gao. "Temperature Controller of Heating Furnace Based on Fuzzy Neural Network Technology." Advanced Materials Research 748 (August 2013): 820–25. http://dx.doi.org/10.4028/www.scientific.net/amr.748.820.
Full textAbstract:
In this study, to solve the problem that heating furnace has the disadvantage of non-linearity, time variant and large delay, a fuzzy neural network controller has been designed according to the combination of fuzzy control and neural networks. In this controller, not only can the reasoning process of neural network be described by the fuzzy rules, but also the fuzzy rules can be dynamically adjusted by the neural network. In addition, the learning algorithm of the fuzzy neural network controller is studied. Simulation results show that the fuzzy neural network controller has good regulating performance and it can meet the needs of heating furnace during industrial production.
16
Ly, Trinh Thi Khanh, Nguyen Thi Thanh, Hoang Thien, and Thai Nguyen. "A Neural Network Controller Design for the Mecanum Wheel Mobile Robot." Engineering, Technology & Applied Science Research 13, no. 2 (2023): 10541–47. http://dx.doi.org/10.48084/etasr.5761.
Full textAbstract:
Advanced controllers are an excellent choice for the trajectory tracking problem of Wheeled Mobile Robots (WMRs). However, these controllers pose a challenge to the hardware structure of WMRs due to the controller's complex structure and the large number of calculations needed. In that context, designing a controller with a simple structure and a small number of computations but good real-time performance is necessary in order to improve the tracking accuracy for the WMRs without requiring high hardware architecture. In this work, a neural network controller with a simple structure for the trajectory-tracking of a Mecanum-Wheel Mobile robot (MWMR) based on a reference controller is proposed. A two-layer feedforward neural network is designed as a tracking controller for the robot. The neural network is trained with a sample input-output data set so that the error between the neural network output and the reference control signal of the supervisory controller is minimal. The neural network parameters are trained to update over time. The simulation results verified the effectiveness of the neural network controller, whose parameters are tuned adaptively to ensure a fast convergence to the desired Bézier trajectory.
17
Joshi, Girisha, and Pinto Pius A J. "ANFIS controller for vector control of three phase induction motor." Indonesian Journal of Electrical Engineering and Computer Science 19, no. 3 (2020): 1177. http://dx.doi.org/10.11591/ijeecs.v19.i3.pp1177-1185.
Full textAbstract:
For variable speed drive applications such as electric vehicles, 3 phase induction motor is used and is controlled by fuzzy logic controllers. For the steady functioning of the vehicle drive, it is essential to generate required torque and speed during starting, coasting, free running, braking and reverse operating regions. The drive performance under these transient conditions are studied and presented. In the present paper, vector control technique is implemented using three fuzzy logic controllers. Separate Fuzzy logic controllers are used to control the direct axis current, quadrature axis current and speed of the motor. In this paper performance of the indirect vector controller containing artificial neural network based fuzzy logic (ANFIS) based control system is studied and compared with regular fuzzy logic system, which is developed without using artificial neural network. Data required to model the artificial neural network based fuzzy inference system is obtained from the PI controlled induction motor system. Results obtained in MATLAB-SIMULINK simulation shows that the ANFIS controller is superior compared to controller which is implemented only using fuzzy logic, under all dynamic conditions.
18
Girisha, Joshi, and Pius A. J. Pinto. "ANFIS controller for vector control of three phase induction motor." Indonesian Journal of Electrical Engineering and Computer Science (IJEECS) 19, no. 3 (2020): 1177–85. https://doi.org/10.11591/ijeecs.v19.i3.pp1177-1185.
Full textAbstract:
For variable speed drive applications such as electric vehicles, 3 phase induction motor is used and is controlled by fuzzy logic controllers. For the steady functioning of the vehicle drive, it is essential to generate required torque and speed during starting, coasting, free running, braking and reverse operating regions. The drive performance under these transient conditions are studied and presented. In the present paper, vector control technique is implemented using three fuzzy logic controllers. Separate Fuzzy logic controllers are used to control the direct axis current, quadrature axis current and speed of the motor. In this paper performance of the indirect vector controller containing artificial neural network based fuzzy logic (ANFIS) based control system is studied and compared with regular fuzzy logic system, which is developed without using artificial neural network. Data required to model the artificial neural network based fuzzy inference system is obtained from the PI controlled induction motor system. Results obtained in MATLABSIMULINK simulation shows that the ANFIS controller is superior compared to controller which is implemented only using fuzzy logic, under all dynamic conditions.
19
Zhang, Wen Hui. "Adaptive Neural Network Control for X-Y NC Position Table." Key Engineering Materials 589-590 (October 2013): 654–57. http://dx.doi.org/10.4028/www.scientific.net/kem.589-590.654.
Full textAbstract:
An adaptive neural network control scheme for X-Y position table is proposed by the paper. X-Y position table model is established, controller based on neutral network is used to learn adaptive and compensate uncertainty model of X-Y position table, neutral network is used to study model. Neural network parameters base on stochastic gradient algorithm can be adjusted adaptive on line. Neural network controller base on augmented variable method is designed.
20
Puentes, Kevin, Luis Morales, David F. Pozo-Espin, and Viviana Moya. "Enhancing Control Systems with Neural Network-Based Intelligent Controllers." Emerging Science Journal 8, no. 4 (2024): 1243–61. http://dx.doi.org/10.28991/esj-2024-08-04-01.
Full textAbstract:
The primary challenge faced by a neural controller in the dynamic model of a mobile robot lies in its ability to address the inherent complexity of the system dynamics. Given that mobile robots exhibit nonlinear movements and are subject to diverse environmental conditions, they contend with a challenging dynamic environment. The neural controllers must demonstrate the capability to continuously adapt and effectively learn to manage the variability present in the dynamic of the robot. This paper presents two intelligent controllers utilizing neural networks, showcasing their relevance in the field of robotics. The first controller, referred to as the neural PID (PIDN), integrates the traditional PID controller with a neural component. The second controller leverages the dynamic model of a differential robot to improve trajectory tracking, employing a parallel architecture that combines PID with neural networks (PID+NN). Our proposals adhere to a cascading structure, where the outer loop takes the lead in reducing position errors through a kinematic controller, while concurrently, the inner loop is employed to regulate linear and angular velocities through the proposed controllers. The controllers are validated in the CoppeliaSIM simulator, offering a realistic setting for evaluating the behavior of the chosen Pioneer 3-DX robot. To comprehensively assess controller performance, three strategies are examined: PIDN, PID+NN, and the conventional PID. Through a blend of qualitative and quantitative analyses, employing diverse performance metrics, the advantages of our proposed controllers become apparent. Doi: 10.28991/ESJ-2024-08-04-01 Full Text: PDF
21
Han, Qiang, Farid Boussaid, and Mohammed Bennamoun. "Soft Fuzzy Reinforcement Neural Network Proportional–Derivative Controller." Applied Sciences 15, no. 9 (2025): 5071. https://doi.org/10.3390/app15095071.
Full textAbstract:
Controlling systems with highly nonlinear or uncertain dynamics present significant challenges, particularly when using conventional Proportional–Integral–Derivative (PID) controllers, as they can be difficult to tune. While PID controllers can be adapted for such systems using advanced tuning methods, they often struggle with lag and instability due to their integral action. In contrast, fuzzy Proportional–Derivative (PD) controllers offer a more responsive alternative by eliminating reliance on error accumulation and enabling rule-based adaptability. However, their industrial adoption remains limited due to challenges in manual rule design. To overcome this limitation, Fuzzy Neural Networks (FNNs) integrate neural networks with fuzzy logic, enabling self-learning and reducing reliance on manually crafted rules. However, most fuzzy neural network PD (FNNPD) controllers rely on mean square error (MSE)-based training, which can be inefficient and unstable in complex, dynamic systems. To address these challenges, this paper presents a Soft Fuzzy Reinforcement Neural Network PD (SFPD) controller, integrating the Soft Actor–Critic (SAC) framework into FNNPD control to improve training speed and stability. While the actor–critic framework is widely used in reinforcement learning, its application to FNNPD controllers has been unexplored. The proposed controller leverages reinforcement learning to autonomously adjust parameters, eliminating the need for manual tuning. Additionally, entropy-regularized stochastic exploration enhances learning efficiency. It can operate with or without expert knowledge, leveraging neural network-driven adaptation. While expert input is not required, its inclusion accelerates convergence and improves initial performance. Experimental results show that the proposed SFPD controller achieves fast learning, superior control performance, and strong robustness to noise, making it effective for complex control tasks.
22
Woodford, Grant W., and Mathys C. du Plessis. "Complex Morphology Neural Network Simulation in Evolutionary Robotics." Robotica 38, no. 5 (2019): 886–902. http://dx.doi.org/10.1017/s0263574719001140.
Full textAbstract:
SUMMARYThis paper investigates artificial neural network (ANN)-based simulators as an alternative to physics-based approaches for evolving controllers in simulation for a complex snake-like robot. Prior research has been limited to robots or controllers that are relatively simple. Benchmarks are performed in order to identify effective simulator topologies. Additionally, various controller evolution strategies are proposed, investigated and compared. Using ANN-based simulators for controller fitness estimation during controller evolution is demonstrated to be a viable approach for the high-dimensional problem specified in this work.
23
Darsivan, Fadly Jashi, Wahyudi Martono, and Waleed F. Faris. "Active Engine Mounting Control Algorithm Using Neural Network." Shock and Vibration 16, no. 4 (2009): 417–37. http://dx.doi.org/10.1155/2009/257480.
Full textAbstract:
This paper proposes the application of neural network as a controller to isolate engine vibration in an active engine mounting system. It has been shown that the NARMA-L2 neurocontroller has the ability to reject disturbances from a plant. The disturbance is assumed to be both impulse and sinusoidal disturbances that are induced by the engine. The performance of the neural network controller is compared with conventional PD and PID controllers tuned using Ziegler-Nichols. From the result simulated the neural network controller has shown better ability to isolate the engine vibration than the conventional controllers.
24
Beham, M., and D. L. Yu. "On-line control for optimal ignition timing using the pseudolinear radial basis function and the local linear model tree." Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering 219, no. 2 (2005): 227–40. http://dx.doi.org/10.1243/095440705x6712.
Full textAbstract:
A new generation of engines demands new control strategies. The increased number of control variables of variable valve timing engines results in complexity of conventional control structures. This necessitates the integration of new technologies for optimal control of the ignition timing. This paper presents a neural network controller for ignition timing that uses two recently proposed new neural network structures—a pseudolinear radial basis function (PLRBF) network and a local linear model tree (LOLIMOT) network. Tests showed that the relative load signal is not necessary to evaluate the ignition angle, and therefore no air mass meter is necessary. The two neural networks are compared with a conventional look-up table control structure. The network controller improves the conventional look-up table method for calibration by comparison with bilateral look-up tables. The neural controller is implemented and tested in a research car. Experimental results show that the neural networks are very effective in mapping non-linearity. The design of the neural network controller simplifies the structure drastically.
25
K.J., Rathi, and S. Ali M. "Neural Network Controller for Power Electronics Circuits." IAES International Journal of Artificial Intelligence (IJ-AI) 6, no. 2 (2017): 49–55. https://doi.org/10.5281/zenodo.4108248.
Full textAbstract:
Artificial Intelligence (AI) techniques, particularly the neural networks, are recently having significant impact on power electronics. This paper explores the perspective of neural network applications in the intelligent control for power electronics circuits. The Neural Network Controller (NNC) is designed to track the output voltage and to improve the performance of power electronics circuits. The controller is designed and simulated using MATLAB-SIMULINK
26
Ji, H., and Zhi Yong Li. "Study on Intelligent Controller of Permanent Magnet Synchronous Motor." Advanced Materials Research 764 (September 2013): 149–53. http://dx.doi.org/10.4028/www.scientific.net/amr.764.149.
Full textAbstract:
This paper puts forward a novel design method of controller based on BP neural network, which is applied to the permanent magnet synchronous motor (PMSM) double closed loop speed regulation system of speed regulator, by using the neural network controller instead of traditional PID controller. It applies the nonlinear adaptive ability of neural network for optimizing the control parameters of PID controller for PMSM. The simulation model was established in Matlab/Simulink. The simulation results indicate that the neutral network PID controller, compared with the traditional PID, has strong robustness and adaptive ability to the model and environments, indicating the good dynamic and static characteristics and control effects.
27
Hu, Guan Shan. "Adaptive Control Based on Neural Network for Ship Sterring Autopilot." Advanced Materials Research 503-504 (April 2012): 1239–42. http://dx.doi.org/10.4028/www.scientific.net/amr.503-504.1239.
Full textAbstract:
The Autopilot is importance for a ship to navigate safely and economically, so we proposes an intelligent reference modeling adaptive controller for ship steering based on neural networks. In order to satisfy the requirements of ship’s course control under various sea status, we used fuzzy logic and neural networks to design the feedback controller, used multilayer perceptron neural network to design the reference model and the identification network. In order to enhance adaptive characteristics of the controller，the parameters of membership functions and connection weights etc were revised online with neural network learning algorithm. The results of simulation shown that the performance of the ship controller is valuable and effective.
28
Obed, Adel, and Ameer Saleh. "Speed Control of BLDC Motor Based on Recurrent Wavelet Neural Network." Iraqi Journal for Electrical and Electronic Engineering 10, no. 2 (2014): 118–29. http://dx.doi.org/10.37917/ijeee.10.2.7.
Full textAbstract:
In recent years, artificial intelligence techniques such as wavelet neural network have been applied to control the speed of the BLDC motor drive. The BLDC motor is a multivariable and nonlinear system due to variations in stator resistance and moment of inertia. Therefore, it is not easy to obtain a good performance by applying conventional PID controller. The Recurrent Wavelet Neural Network (RWNN) is proposed, in this paper, with PID controller in parallel to produce a modified controller called RWNN-PID controller, which combines the capability of the artificial neural networks for learning from the BLDC motor drive and the capability of wavelet decomposition for identification and control of dynamic system and also having the ability of self-learning and self-adapting. The proposed controller is applied for controlling the speed of BLDC motor which provides a better performance than using conventional controllers with a wide range of speed. The parameters of the proposed controller are optimized using Particle Swarm Optimization (PSO) algorithm. The BLDC motor drive with RWNN-PID controller through simulation results proves a better in the performance and stability compared with using conventional PID and classical WNN-PID controllers.
29
Psaltis, D., A. Sideris, and A. A. Yamamura. "A multilayered neural network controller." IEEE Control Systems Magazine 8, no. 2 (1988): 17–21. http://dx.doi.org/10.1109/37.1868.
Full text
30
PARK, JU H., and O. M. KWON. "SYNCHRONIZATION OF NEURAL NETWORKS OF NEUTRAL TYPE WITH STOCHASTIC PERTURBATION." Modern Physics Letters B 23, no. 14 (2009): 1743–51. http://dx.doi.org/10.1142/s0217984909019909.
Full textAbstract:
In this letter, the problem of feedback controller design to achieve synchronization for neural network of neutral type with stochastic perturbation is considered. Based on Lyapunov method and LMI (linear matrix inequality) framework, the goal of this letter is to derive an existence criterion of the controller for the synchronization between master and response networks.
31
Majeed, Anwar Dhyaa, and Nadia Adnan Shiltagh Al-Jamali. "Spike neural network as a controller in SDN network." Journal of Engineering 27, no. 9 (2021): 64–77. http://dx.doi.org/10.31026/j.eng.2021.09.06.
Full textAbstract:
The paper proposes a methodology for predicting packet flow at the data plane in smart SDN based on the intelligent controller of spike neural networks(SNN). This methodology is applied to predict the subsequent step of the packet flow, consequently reducing the overcrowding that might happen. The centralized controller acts as a reactive controller for managing the clustering head process in the Software Defined Network data layer in the proposed model. The simulation results show the capability of Spike Neural Network controller in SDN control layer to improve the (QoS) in the whole network in terms of minimizing the packet loss ratio and increased the buffer utilization ratio.
32
Alyukov, Alexander, Yuri Rozhdestvenskiy, and Sergei Aliukov. "Active Shock Absorber Control Based on Time-Delay Neural Network." Energies 13, no. 5 (2020): 1091. http://dx.doi.org/10.3390/en13051091.
Full textAbstract:
A controlled suspension usually consists of a high-level and a low-level controller. The purpose the high-level controller is to analyze external data on vehicle conditions and make decisions on the required value of the force on the shock absorber rod, while the purpose of the low-level controller is to ensure the implementation of the desired force value by controlling the actuators. Many works have focused on the design of high-level controllers of active suspensions, in which it is considered that the shock absorber can instantly and absolutely accurately implement a given control input. However, active shock absorbers are complex systems that have hysteresis. In addition, electro-pneumatic and hydraulic elements are often used in the design, which have a long response time and often low accuracy. The application of methods of control theory in such systems is often difficult due to the complexity of constructing their mathematical models. In this article, the authors propose an effective low-level controller for an active shock absorber based on a time-delay neural network. Neural networks in this case show good learning ability. The low-level controller is implemented in a simplified suspension model and the simulation results are presented for a number of typical cases.
33
Elgargani, T. N., A. A. Hudoud, and S. O. Abid. "IMPROVEMENT AND CONTROL OF THE SPEED RESPONSE OF THE PERMANENT MAGNET SYNCHRONOUS MOTOR DRIVE USING A FUZZY – PI CONTROLLER." Journal of Science and Technology 30, no. 5 (2025): 25–36. https://doi.org/10.20428/jst.v30i5.2812.
Full textAbstract:
High-speed and high-performance electric motors are designed to reach a high level of demand control. The permanent magnet synchronous motors (PMSMs) drive has a non-linear model that is not easy to deal with using traditional control methods when controlling the three phase motors because of their nature, (intricate highly non-linear model). Therefore, neural networks controllers compared with fuzzy logic controllers (FLCs) are getting more attention among researchers, as they can be used for such systems. The neural networks controller relies on training of this mathematical model, and the fuzzy controller also relies on experience. The performance of these two controllers were compared to each other in terms of output response. As all the real systems exhibit non-linear behavior, conventional PI (Proportional-Integral) controllers are unable to provide good and acceptable results. For this reason, when designing intelligent control systems, the corresponding model for simulation should reflect all characteristics of the real system to be controlled. The basic idea of this paper is to apply the fuzzy-PI controller on PMSMs drive and compare the obtained results with the traditional PI. Also, one intelligent controller, which is the NN (Neural Network) controller, is applied and its performance is simulated and studied. MATLAB/SIMULINK environment is used for design, implementation and testing. Therefore, the speed and torque of the PMSMs drive can be controlled satisfactorily. Finally, simulation results have shown decent results in the improvement of the system behavior.
34
Shi, Yong, Lian Yu Zhang, Jun Sun, and Hong Guang Zhang. "Research on the Speed of Diesel Engine Based on Improved BP Neural Network Controller." Applied Mechanics and Materials 281 (January 2013): 105–11. http://dx.doi.org/10.4028/www.scientific.net/amm.281.105.
Full textAbstract:
Marine diesel engine is of characteristics of non-linear and time-invariant, so it is difficult to be controlled with traditional PID controller. An adaptive controller based on back-propagation (BP) neural networks was put forwarded for marine diesel engine speed control system, where two neural networks are proposed to control the position loop and speed loop. The adaptive controller was improved was improved via introducing relative error in target evaluation function of the BP neural network, and obtain sensitivity function of diesel engine output with respect to its input using a differential equation. The controller has self-learning and adaptive capacity. It can also optimize the PID controller parameters online. The controller was experimentally evaluated on rack position actuator of marine diesel engine simulated based on a diesel hardware-in-loop system of dSPACE. Finally, tests on a diesel engine demonstrated that the controller can satisfy the transient and steady demands of speed regulation system.
35
Rathi, K. J., and M. S. Ali. "Neural Network Controller for Power Electronics Circuits." IAES International Journal of Artificial Intelligence (IJ-AI) 6, no. 2 (2017): 49. http://dx.doi.org/10.11591/ijai.v6.i2.pp49-55.
Full textAbstract:
Artificial Intelligence (AI) techniques, particularly the neural networks, are recently having significant impact on power electronics. This paper explores the perspective of neural network applications in the intelligent control for power electronics circuits. The Neural Network Controller (NNC) is designed to track the output voltage and to improve the performance of power electronics circuits. The controller is designed and simulated using MATLAB-SIMULINK
36
Sathishkumar, H., та S. S. Parthasarathy. "Identification of robust controller for 3hp 3Φ induction motor". IAES International Journal of Robotics and Automation (IJRA) 8, № 2 (2019): 125–32. https://doi.org/10.11591/ijra.v8i2.pp125-132.
Full textAbstract:
This paper deals about the identification of robust controller for 3hp 3Φ induction motor which is used in cable industry (Ravicab cables private limited) at Bidadi. In this cable industry 3hp 3Φ induction motor is used for cable pulling purpose. This industry is using PID (Proportional derivative integral) controller based VFD (Voltage frequency drive) for controlling the speed of this 3hp 3Φ induction motor. This VFD is not functioning well for the non linear load and disturbance environment. Therefore in this paper Neural network based speed controller is proposed as proposed controller-I for replacing the PID based VFD. Performance of the 3hp 3Φ induction motor is estimated when Neural network controller is interfaced with the motor. Then Neuro-fuzzy controller based speed controller is proposed as proposed controller-II for replacing the PID based VFD. Performance of the 3hp 3Φ induction motor is estimated when Neuro-fuzzy controller is interfaced with the motor. At last robust controller for the 3hp 3Φ induction motor which is used for cable pulling purpose is going to be identified by doing comparison chart between Neural network and Neuro-fuzzy controller.
37
Somwong, Poom, Karn Patanukhom, and Yuthapong Somchit. "Energy-Aware Controller Load Distribution in Software-Defined Networking using Unsupervised Artificial Neural Networks." Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications 16, no. 1 (2025): 289–314. https://doi.org/10.58346/jowua.2025.i1.018.
Full textAbstract:
Software-Defined Networking (SDN) enhances network management by separating the control and data planes into controllers and switches, allowing for centralized, programmable networks with multiple controllers. Switches are mapped to controllers and exchange control messages to manage the network, which leads to significant energy consumption. Managing energy in networks has become a critical issue, as dynamic changes in switch loads can cause controller overloads, necessitating the migration of switches to other controllers. As networks grow, energy consumed in control communications becomes a major concern. This paper proposes an unsupervised learning Artificial Neural Network (ANN) model to address controller overloads and optimize energy consumption, achieving faster execution times compared to conventional methods while maintaining manageable energy efficiency. The model considers dynamic switch loads and the hop distance between switches and controllers when remapping switches to optimize energy use. Experimental results demonstrate that the proposed unsupervised ANN model performs effectively in large networks, enabling efficient handling of controller overloads during variations in switch loads. The adaptability of the ANN model provides a robust strategy for energy-efficient load distribution, enhancing the scalability and efficiency of SDN environments.
38
Yu, Jinlong, Zhi Li, Lu Jia, and Yasheng Zhang. "Switching Neural Network Control for Underactuated Spacecraft Formation Reconfiguration in Elliptic Orbits." Applied Sciences 12, no. 12 (2022): 5792. http://dx.doi.org/10.3390/app12125792.
Full textAbstract:
A switching neural network control scheme, consisting of the adaptive neural network controller and sliding mode controller, is proposed for underactuated formation reconfiguration in elliptic orbits with the loss of either the radial or in-track thrust. By using the inherent coupling of system states, the switching neural network technique is then adopted to estimate the unmatched disturbances and design the underactuated controller to achieve underactuated formation reconfiguration with high precision. The adaptive neural network controller works in the active region, and the disturbances composed of linearization errors and external perturbations are approximated by radial basis function neural networks. The adaptive sliding mode controller works outside the active region, and the upper bound of the approximation errors is estimated by the adaptation law. The stability of the closed-loop control system is proved via the Lyapunov-based approach. The numerical simulation results have demonstrated the rapid, high-precision and robust performance of the proposed controller compared with the linear sliding mode controller.
39
Voevoda, Alexsander, and Victor Shipagin. "Synthesis of a neural network control regulator of a nonlinear model of an inverted pendulum on a cart." Science Bulletin of the Novosibirsk State Technical University, no. 2-3 (November 13, 2020): 25–36. http://dx.doi.org/10.17212/1814-1196-2020-2-3-25-36.
Full textAbstract:
In this article, we consider a method for selecting a structure of a neural network used to regulate an "inverted pendulum on a cart" object taking into account its additional features of a mathematical description, namely, nonlinear parameters. The algorithm is illustrated by the example of control synthesis which includes two neuroregulators. One of them is responsible for bringing the cart to the specified position, and the second is responsible for holding the pendulum in a vertical position. The structure transformations will be performed for the controller responsible for bringing the cart to the specified position. The architecture of a neural network controller is based on a discrete controller synthesized using polynomial matrix decomposition. For the original controller, we define the limits of its possible control of a nonlinear system. To increase the range of control of a nonlinear object, we perform transformations of the neural network structure of the original controller. We will make some complications in the structure of the neural network of the regulator, namely, increase the number of neurons and replace some activation functions with nonlinear ones (hyperbolic tangent). Next, we suggest one of the ways to select initial values of weight coefficients. Then we train the neural network and check the performance of the resulting controller on a nonlinear object. At the next stage, we compare the obtained performance of a controller having a complicated neural network structure with the performance of a classical controller. Thus, the purpose of this study is to formalize the synthesis procedure for a neural network controller for controlling a nonlinear object using a calculated classical controller for a linearized object model. The proposed method of generating the architecture of a neural network of controllers makes it possible to increase the range of control by a nonlinear object in comparison with the controller obtained by the method of polynomial matrix decomposition for a linear object. Compared to the typical ones, the proposed neural network structure is not redundant and therefore does not require additional computing resources to configure it.
40
Machavarapu, Suman, Mannam Venu Gopala Rao, and Pulipaka Venkata Ramana Rao. "Design of Load Frequency Controller for Multi-area System Using AI Techniques." Journal Européen des Systèmes Automatisés 53, no. 4 (2020): 541–48. http://dx.doi.org/10.18280/jesa.530413.
Full textAbstract:
The paper presents an adaptive Load Frequency Controller (LFC) based on a neural network for the interconnected multi-area systems. When there is an imbalance between active power generation and demand there will deviation in the frequency from the reference value. Major disturbances that lead to the variation in frequency beyond the allowable limits are variation in load demand and faults, etc. Initially PID based LFC which is a conventional controller is used to bring back the variations in frequency when there is a disturbance. But these conventional controllers will operate certain operating points only, very slow and, are less efficient for nonlinear systems. To avoid the flaws in the conventional controller the artificial intelligent controllers such as neural network and fuzzy logic controllers are designed. The three, two area, and single area systems are considered as the test systems. The response of all the test systems is observed without and with PI, fuzzy, and neural network controllers. It was observed that the neural network controller is outperforming in damping the variation in the frequency due to the disturbances.
41
Asgari, Hamid, Mohsen Fathi Jegarkandi, XiaoQi Chen, and Raazesh Sainudiin. "Design of conventional and neural network based controllers for a single-shaft gas turbine." Aircraft Engineering and Aerospace Technology 89, no. 1 (2017): 52–65. http://dx.doi.org/10.1108/aeat-11-2014-0187.
Full textAbstract:
Purpose The purpose of this paper is to develop and compare conventional and neural network-based controllers for gas turbines. Design/methodology/approach Design of two different controllers is considered. These controllers consist of a NARMA-L2 which is an artificial neural network-based nonlinear autoregressive moving average (NARMA) controller with feedback linearization, and a conventional proportional-integrator-derivative (PID) controller for a low-power aero gas turbine. They are briefly described and their parameters are adjusted and tuned in Simulink-MATLAB environment according to the requirement of the gas turbine system and the control objectives. For this purpose, Simulink and neural network-based modelling is used. Performances of the controllers are explored and compared on the base of design criteria and performance indices. Findings It is shown that NARMA-L2, as a neural network-based controller, has a superior performance to PID controller. Practical implications This study aims at using artificial intelligence in gas turbine control systems. Originality/value This paper provides a novel methodology for control of gas turbines.
42
Mahdi, Shaymaa Mahmood, and Omar Farouq Lutfy. "Control of a servo-hydraulic system utilizing an extended wavelet functional link neural network based on sine cosine algorithms." Indonesian Journal of Electrical Engineering and Computer Science 25, no. 2 (2022): 847–56. https://doi.org/10.11591/ijeecs.v25.i2.pp847-856.
Full textAbstract:
Servo-hydraulic systems have been extensively employed in various industrial applications. However, these systems are characterized by their highly complex and nonlinear dynamics, which complicates the control design stage of such systems. In this paper, an extended wavelet functional link neural network (EWFLNN) is proposed to control the displacement response of the servo-hydraulic system. To optimize the controller's parameters, a recently developed optimization technique, which is called the modified sine cosine algorithm (M-SCA), is exploited as the training method. The proposed controller has achieved remarkable results in terms of tracking two different displacement signals and handling external disturbances. From a comparative study, the proposed EWFLNN controller has attained the best control precision compared with those of other controllers, namely, a proportional-integralderivative (PID) controller, an artificial neural network (ANN) controller, a wavelet neural network (WNN) controller, and the original wavelet functional link neural network (WFLNN) controller. Moreover, compared to the genetic algorithm (GA) and the original sine cosine algorithm (SCA), the M-SCA has shown better optimization results in finding the optimal values of the controller's parameters.
43
Sun, Qi-Ming, and Hong-Sen Yan. "Multidimensional Taylor Network Optimal Control of MIMO Nonlinear Systems without Models for Tracking by Output Feedback." Mathematical Problems in Engineering 2017 (2017): 1–9. http://dx.doi.org/10.1155/2017/1548095.
Full textAbstract:
The actual controlled objects are generally multi-input and multioutput (MIMO) nonlinear systems with imprecise models or even without models, so it is one of the hot topics in the control theory. Due to the complex internal structure, the general control methods without models tend to be based on neural networks. However, the neuron of neural networks includes the exponential function, which contributes to the complexity of calculation, making the neural network control unable to meet the real-time requirements. The newly developed multidimensional Taylor network (MTN) requires only addition and multiplication, so it is easy to realize real-time control. In the present study, the MTN approach is extended to MIMO nonlinear systems without models to realize adaptive output feedback control. The MTN controller is proposed to guarantee the stability of the closed-loop system. Our experimental results show that the output signals of the system are bounded and the tracking error goes nearly to zero. The MTN optimal controller is proven to promise far better real-time dynamic performance and robustness than the BP neural network self-adaption reconstitution controller.
44
Pletl, Szilveszter, and Bela Lantos. "Advanced Robot Control Algorithms Based on Fuzzy, Neural and Genetic Methods." Journal of Advanced Computational Intelligence and Intelligent Informatics 5, no. 2 (2001): 81–89. http://dx.doi.org/10.20965/jaciii.2001.p0081.
Full textAbstract:
Soft computing (fuzzy systems, neural networks and genetic algorithms) can solve difficult problems, especially non-linear control problems such as robot control. In the paper two algorithms have been presented for the nonlinear control of robots. The first algorithm applies a new neural network based controller structure and a learning method with stability guarantee. The controller consists of the nonlinear prefilter, the feedforward neural network and feadback PD controllers. The fast learning algorithm of the neural network is based on Moore-Penrose pseudoinverse technique. The second algorithm is based on a decentralized hierarchical neuro-fuzzy controller structure. New approach to evolutionary algorithms called LEGA optimizes the controller during the teaching period. LEGA combines the standard GA technique with numerical optimum seeking for a limited number of elite individuels in each generation. It can lead to global optimum in few generations. The soft computing based nonlinear control algorithms have been applied for the control of a rigid link flexible joint (RLFJ) 4 DOF SCARA robot in order to prove the effectiveness of the proposed methods.
45
Ivanov, Radoslav, Kishor Jothimurugan, Steve Hsu, Shaan Vaidya, Rajeev Alur, and Osbert Bastani. "Compositional Learning and Verification of Neural Network Controllers." ACM Transactions on Embedded Computing Systems 20, no. 5s (2021): 1–26. http://dx.doi.org/10.1145/3477023.
Full textAbstract:
Recent advances in deep learning have enabled data-driven controller design for autonomous systems. However, verifying safety of such controllers, which are often hard-to-analyze neural networks, remains a challenge. Inspired by compositional strategies for program verification, we propose a framework for compositional learning and verification of neural network controllers. Our approach is to decompose the task (e.g., car navigation) into a sequence of subtasks (e.g., segments of the track), each corresponding to a different mode of the system (e.g., go straight or turn). Then, we learn a separate controller for each mode, and verify correctness by proving that (i) each controller is correct within its mode, and (ii) transitions between modes are correct. This compositional strategy not only improves scalability of both learning and verification, but also enables our approach to verify correctness for arbitrary compositions of the subtasks. To handle partial observability (e.g., LiDAR), we additionally learn and verify a mode predictor that predicts which controller to use. Finally, our framework also incorporates an algorithm that, given a set of controllers, automatically synthesizes the pre- and postconditions required by our verification procedure. We validate our approach in a case study on a simulation model of the F1/10 autonomous car, a system that poses challenges for existing verification tools due to both its reliance on LiDAR observations, as well as the need to prove safety for complex track geometries. We leverage our framework to learn and verify a controller that safely completes any track consisting of an arbitrary sequence of five kinds of track segments.
46
Li, Yun, Yufei Wu, Xiaohui Zhang, Xinglin Tan, and Wei Zhou. "Unmanned Bicycle Balance Control Based on Tunicate Swarm Algorithm Optimized BP Neural Network PID." International Journal of Information Technologies and Systems Approach 16, no. 3 (2023): 1–16. http://dx.doi.org/10.4018/ijitsa.324718.
Full textAbstract:
In this study, the authors introduce a novel approach that leverages the tunicate swarm algorithm (TSA) to optimize proportional-integral-derivative (PID) controller based on a back propagation (BP) neural network. The core objective of the approach is to manage and counteract uncertainties and disturbance that may jeopardize the balance and stability of self-driving bicycles in operation. By using the self-learning capabilities of BP neural networks, the controller can dynamically adjust PID parameters in real time. This enables an enhanced robustness and reliability during operation. Further bolstering the efficiency of our controller, the authors use the TSA to optimize the initial weights of a neural network. This effectively mitigates the commonly associated with slow convergence and being entrapped in local minima. Through simulation and experimentation, the findings reveal that the TSA-optimized BP neural network PID controller dramatically improves dynamic performance and robustness. It also proficiently manages changes in the environment such as wind and ground bumps. Therefore, the proposed controller design offers an effective solution to the balancing problem of self-driving bicycles and paves the way for a promising future in designing versatile controllers with broad application potential.
47
Scott, Gary M., Jude W. Shavlik, and W. Harmon Ray. "Refining PID Controllers Using Neural Networks." Neural Computation 4, no. 5 (1992): 746–57. http://dx.doi.org/10.1162/neco.1992.4.5.746.
Full textAbstract:
The KBANN (Knowledge-Based Artificial Neural Networks) approach uses neural networks to refine knowledge that can be written in the form of simple propositional rules. We extend this idea further by presenting the MANNCON (Multivariable Artificial Neural Network Control) algorithm by which the mathematical equations governing a PID (Proportional-Integral-Derivative) controller determine the topology and initial weights of a network, which is further trained using backpropagation. We apply this method to the task of controlling the outflow and temperature of a water tank, producing statistically significant gains in accuracy over both a standard neural network approach and a nonlearning PID controller. Furthermore, using the PID knowledge to initialize the weights of the network produces statistically less variation in test set accuracy when compared to networks initialized with small random numbers.
48
Jafari, Mohammad, and Hao Xu. "Intelligent Control for Unmanned Aerial Systems with System Uncertainties and Disturbances Using Artificial Neural Network." Drones 2, no. 3 (2018): 30. http://dx.doi.org/10.3390/drones2030030.
Full textAbstract:
Stabilizing the Unmanned Aircraft Systems (UAS) under complex environment including system uncertainties, unknown noise and/or disturbance is so challenging. Therefore, this paper proposes an adaptive neural network based intelligent control method to overcome these challenges. Based on a class of artificial neural network, named Radial Basis Function (RBF) networks an adaptive neural network controller is designed. To handle the unknown dynamics and uncertainties in the system, firstly, we develop a neural network based identifier. Then, a neural network based controller is generated based on both the identified model of the system and the linear or nonlinear controller. To ensure the stability of the system during its online training phase, the linear or nonlinear controller is utilized. The learning capability of the proposed intelligent controller makes it a promising approach to take system uncertainties, noises and/or disturbances into account. The satisfactory performance of the proposed intelligent controller is validated based on the computer based simulation results of a benchmark UAS with system uncertainties and disturbances, such as wind gusts disturbance.
49
Jin, Fu, Jian Jun Sun, and Hong Bin Yu. "Design of Control System of Eddy Current Retarder Based on BP Neural Network PID Controller." Applied Mechanics and Materials 494-495 (February 2014): 223–28. http://dx.doi.org/10.4028/www.scientific.net/amm.494-495.223.
Full textAbstract:
A new kind of algorithm of controller for eddy current retarder is designed in this paper. The eddy current retarder control system with traditional PID controller can't achieve a perfect performance in the rapid response. Back propagation (BP) neural network is one of artificial neural networks which has a good learning ability with a simple and recurrent structure, so it is suitable for controlling complicated eddy current retarder system. This paper introduces the principle, characteristics and learning algorithm of the BP neural network and designs the control system of eddy current retarder based on BP neural network PID controller by combining BP neural network and traditional PID. Making use of MATLAB, simulate this new kind of controller for eddy current retarder in the rapid response. Simulation results show it can improve the dynamic response performance and enhance the static precision compared to the traditional PID controller.
50
Talebi, H. A., K. Khorasani, and R. V. Patel. "Tracking control of a flexible-link manipulator using neural networks: experimental results." Robotica 20, no. 4 (2002): 417–27. http://dx.doi.org/10.1017/s026357470200406x.
Full textAbstract:
In this paper, the problem of tip position tracking control of a flexible-link manipulator is considered. Two neural network schemes are presented. In the first scheme, the controller is composed of a stabilizing joint PD controller and a neural network tracking controller. The objective is to simultaneously achieve hub-position tracking and control of the elastic deflections at the tip. In the second scheme, tracking control of a point along the arm is considered to avoid difficulties associated with the output feedback control of a non-minimum phase flexible manipulator. A separate neural network is employed for determining an appropriate output to be used for feedback. The controller is also composed of a neural network tracking controller and a stabilizing joint PD controller. Experimental results on a single-link flexible manipulator show that the proposed networks result in significant improvements in the system response with an increase in controller dynamic range despite changes in the desired trajectory.