Search

This article presents an intelligent control strategy for the position and attitude tracking of a quadrotor using a nonsingular fast-terminal sliding mode and disturbance observer. The quadrotor system is divided into an underactuated position and the fully-actuated attitude subsystem. A single hidden layer feed-forward neural network is used for the estimation of unknown dynamics associated with the attitude subsystem. Moreover, a nonlinear disturbance observer is designed to tackle the unknown bounded external disturbances acting on the attitude subsystem by supplying its estimation in the control laws that help in mitigating the chattering phenomena. The proposed methodology ensures the finite-time convergence of the tracking error and does not prone to the singularity problem in control. The closed-loop finite-time stability is investigated using the Lyapunov theory. Besides, an expression for the convergence time has been derived. The effectiveness of the proposed scheme is initially assessed using numerical simulations for the way-point tracking as well as circle tracking tasks and then validated in real-time using DJI Matrice 100 by performing a lemniscate tracking task. To show the superiority of the proposed methodology over existing methods, a comparative study has been done and found the improved tracking performance in the proposed approach.

This article investigates hardware implementation of hierarchical temporal memory (HTM), a brain-inspired machine learning algorithm that mimics the key functions of the neocortex and is applicable to many machine learning tasks. Spatial pooler (SP) is one of the main parts of HTM, designed to learn the spatial information and obtain the sparse distributed representations (SDRs) of input patterns. The other part is temporal memory (TM) which aims to learn the temporal information of inputs. The memristor, which is an appropriate synapse emulator for neuromorphic systems, can be used as the synapse in SP and TM circuits. In this article, a memristor-based SP (MSP) circuit structure is designed to accelerate the execution of the SP algorithm. The presented MSP has properties of modeling both the synaptic permanence and the synaptic connection state within a single synapse, and on-device and parallel learning. Simulation results of statistic metrics and classification tasks on several real-world datasets substantiate the validity of MSP.

In this article, a new intelligent hybrid controller is proposed. The controller is based on the combination of the orthogonal endocrine neural network (OENN) and orthogonal endocrine ANFIS (OEANFIS). The orthogonal part of the controller consists of Chebyshev orthogonal functions, which are used because of their recursive property, computational simplicity, and accuracy in nonlinear approximations. Artificial endocrine influence on the controller is achieved by introducing excitatory and inhibitory glands to the OENN part of the structure, in the form of postsynaptic potentials. These potentials provide a network with the capability of additional self-regulation in the presence of external disturbances. The intelligent structure is trained using a developed learning algorithm, which consists of both offline and online learning procedures: online learning for fitting OENN substructure and offline learning for adjusting OEANFIS parameters. The learning process is expanded by introducing the learning rate adaptation algorithm, which bases its calculations on the sign of the error difference. Finally, the proposed intelligent controller was experimentally tested for control of a nonlinear multiple-input-multiple-output two rotor aerodynamical system. During the test phase, an additional four related intelligent control logics and default PID-based controllers were used, and tracking performance comparisons were performed. The proposed controller showed notably better online results in comparison to other control algorithms. The major deficiencies of the structure are complexity and noticeably large training computation time, but these drawbacks can be neglected if tracking performances of a dynamical system are of the highest importance.

Natural gas has been recognized as a promising energy supply for modern society due to its relatively less air pollution in consumption, while pipeline transportation is preferred especially for long-distance transmissions. A simplified pipeline control scenario is proposed in this paper to deeply accelerate the management and decision process in pipeline dispatch, in which a direct relevance between compressor operations and the inlet flux at certain stations is established as the main dispatch logic. A deep neural network is designed with specific input and output features for this scenario and the hyper-parameters are carefully tuned for a better adaptability of this problem. The realistic operation data of two pipelines have been obtained and prepared for learning and testing. The proposed algorithm with the optimized network structure is proved to be effective and reliable in predicting the pipeline operation status, under both the normal operation conditions and abnormal situations. The successful definition of ”ghost compressors” make this algorithm to be the first self-adaptive deep learning algorithm to assist natural gas pipeline intelligent control.

Research on the sixth-generation cellular networks (6G) is gaining huge momentum to achieve ubiquitous wireless connectivity. Connected autonomous vehicles (CAVs) is a critical vertical application for 6G, holding great potentials of improving road safety, road and energy efficiency. However, the stringent service requirements of CAV applications on reliability, latency, and high speed communications will present big challenges to 6G networks. New channel access algorithms and intelligent control schemes for connected vehicles are needed for 6G-supported CAV. In this article, we investigated 6G-supported cooperative driving, which is an advanced driving mode through information sharing and driving coordination. First, we quantify the delay upper bounds of 6G vehicle-to-vehicle (V2V) communications with hybrid communication and channel access technologies. A deep learning neural network is developed and trained for the fast computation of the delay bounds in real-time operations. Then, an intelligent strategy is designed to control the intervehicle distance for cooperative autonomous driving. Furthermore, we propose a Markov chain-based algorithm to predict the parameters of the system states, and also a safe distance mapping method to enable smooth vehicular speed changes. The proposed algorithms are implemented in the AirSim autonomous driving platform. Simulation results show that the proposed algorithms are effective and robust with safe and stable cooperative autonomous driving, which greatly improve the road safety, capacity, and efficiency.

High accuracy of terrain identification is essential for intelligent control of tracked mobile robots. In this article, a learning-based identification framework is proposed to achieve precise driving torque prediction. Experiments under straight-line and steady-state turning maneuvers were conducted to develop terrain identification in three similar terrains. A multiple deep belief networks is applied as the identification layer with three kinds of signal sources. An equivalent weight algorithm with training experience effectively integrates the results from different signal sources to improve the identification accuracy. With the experiment and identification results, a method combining numerical approximation and Gaussian process (GP) is presented to predict driving torque. A combined Gaussian kernel with long- and short-term characteristics is selected to enhance prediction performance. The results from combined signal sources under straight-line maneuvers yield over 98% accuracy, which exceeds that from other sources. The integration algorithm obviously improves the identification accuracy and stability compared with a single signal source. The influence of window length in GP is explored with hyperparameters and results. The performance of torque prediction in different terrains is analyzed with inner and outer tracks. Compared with existing methods, the results validate the effectiveness and superior performance of the proposed framework.

In this paper, stable and adaptive neural network compensators are proposed to control the uncertain permanent magnet synchronous motor (PMSM). Firstly, the overall uncertainties caused by mathematical modelling, parameters variation during operation and external load torque disturbances are modelled. Secondly, a new motion control scheme, where (d-q) current loops are dotted by two on-line tuning neural network compensators (NNCs), is used to compensate these uncertainties. As a result, the speed control loop is processed easily by proportional integral (PI) controller. Stability of the closed-loop system is also designed according to the Lyapunov stability. Compared to classical vector control, the simulations of PMSM system at different speeds including nominal, low and high speed, with and without uncertainties, show the effectiveness of the proposed control scheme.

This article considers a robust intelligent control problem for a class of power-electronic converters via a neuro-fuzzy learning mechanism. First, a terminal sliding-mode control (TSMC) is designed to ensure finite-time error convergence and further enhance the system performance. Meanwhile, a saturation function is utilized in the proposed TSMC. Then, by using type-2 fuzzy neural network (T2FNN) to approximate the developed TSMC, the corresponding adaptive T2FNN controller with online parameter adjustment is established. To enhance the generalization ability for the uncertainties, the recurrent feature-selection algorithm is added into T2FNN. Moreover, the existence of adaptive compensator comprised by upper bound updated law can avoid the impact of the approximation error. Finally, to show the superiorities of the T2FNN controller, it is applied to active power filter, and the simulation and experimental results are compared with the existing literature.

In recent years, the importance of automation systems in the industry has undergone a rapid development process, has started to be understood better and the need for a trained labor force has increased in the field. With the advances automation system, more tasks that are normally performed or controlled by humans are achived by machines or automatic systems. Thus, both faster production and higher production performances are achived by machine while labor is utilized other area where it is less hazardaous to human health. To operate such systems safely and economically, advanced control systems are needed. In this study, the transportation of a load with variable load and height have been automated by Zero Vibration (ZV) input shaping and Neural Network (NN) intelligent control methods in the Matlab/Simulink environment. In addition, the mechanical damping methods have been developed against external forces. It has been shown that this is necessary not only for positioning a load with minimum oscillation but also for balancing it against wind forces. The presented study provides theoretical analysis of the problem, related state of the art industrial applications, the results of the simulation and the experimental studies of the proposed techniques.

In this paper, a hybrid neural-genetic fuzzy system is proposed to control the flow and height of water in the reservoirs of water transfer networks. These controls will avoid probable water wastes in the reservoirs and pressure drops in water distribution networks. The proposed approach combines the artificial neural network, genetic algorithm, and fuzzy inference system to improve the performance of the supervisory control and data acquisition stations through a new control philosophy for instruments and control valves in the reservoirs of the water transfer networks. First, a multi-core artificial neural network model, including a multi-layer perceptron and radial based function, is proposed to forecast the daily consumption of the water in a reservoir. A genetic algorithm is proposed to optimize the parameters of the artificial neural networks. Then, the online height of water in the reservoir and the output of artificial neural networks are used as inputs of a fuzzy inference system to estimate the flow rate of the reservoir inlet. Finally, the estimated inlet flow is translated into the input valve position using a transform control unit supported by a nonlinear autoregressive exogenous model. The proposed approach is applied in the Tehran water transfer network. The results of this study show that the usage of the proposed approach significantly reduces the deviation of the reservoir height from the desired levels.

In this study, the control problem of complex robot system with uncertainties and disturbances is addressed. Fuzzy system-fuzzy neural network-backstepping control (FS-FNN-BSC) system is proposed, which can guarantee the accurate, stable and efficient control. First, the general dynamics model of robot is introduced briefly. Then, the design procedure of backstepping control (BSC) technique is presented, to make the best of the advantages of fuzzy system (FS) and fuzzy neural network (FNN) and compromise the accuracy and efficiency, the FS is adopted to approximate the modeling information, and the FNN is utilized to approximate and predict the non-modeling information, and the FS-FNN-BSC system is constructed. Moreover, based on the Lyapunov stability theorem, the stability of the FS-FNN-BSC is proved. To illustrate the correctness, practicality and generality of the proposed control method, the FS-FNN-BSC system is applied to the series robot (KUKA robot) and the parallel robot (Delta robot). And the superiority of the proposed FS-FNN-BSC strategy is highlighted by quantitative comparison with the existing intelligent control methods.

This article reviews the recent development of adaptive dynamic programming (ADP) with applications in control. First, its applications in optimal regulation are introduced, and some skilled and efficient algorithms are presented. Next, the use of ADP to solve game problems, mainly nonzero-sum game problems, is elaborated. It is followed by applications in large-scale systems. Note that although the functions presented in this article are based on continuous-time systems, various applications of ADP in discrete-time systems are also analyzed. Moreover, in each section, not only some existing techniques are discussed, but also possible directions for future work are pointed out. Finally, some overall prospects for the future are given, followed by conclusions of this article. Through a comprehensive and complete investigation of its applications in many existing fields, this article fully demonstrates that the ADP intelligent control method is promising in today’s artificial intelligence era. Furthermore, it also plays a significant role in promoting economic and social development.

Haptic technology has enormous applications in several fields from medical, military, and in our day-to-day life’s products including video games, smartphones, and smart cities. The Haptic Interface Controller (HIC), a key circuitry for interaction between the user and the virtual world, has two main control issues: stability and transparency. These two issues are complementary to each other i.e. emphasis on one will degrade the other and vice-versa. To address this, intelligent control techniques including Genetic Algorithm (GA), Feed-Forward Neural Network (FFNN), and Fuzzy Logic Control (FLC) have been used in design of the HIC. To ensure the performance in real-time, in system parametric uncertainty and delay have been added while designing the HIC so that a balance could be maintained between the two issues.

This study mainly focuses on the development of a second-order sliding mode-based meta-cognitive fuzzy neural network (MCFNN) control for power quality improvement, whereby an MCFNN is employed to learn modelling uncertainties. First, a second-order sliding mode control (SOSMC) is designed to track reference current for active power filter. Then, an MCFNN system with accurate approximation capability is further investigated to learn the unknown parts in SOSMC. Different from the existing predefined structure approaches, only necessary data can be extracted to adjust the structure and parameters of the networks in MCFNN. Subsequently, the Lyapunov stability analysis is presented to guarantee tracking performance and stability of the closed-loop system. Moreover, the excellent performance of the proposed MCFNN scheme is verified by simulation and experimental studies, and its remarkable characteristics are exhibited in comparison with other intelligent control schemes.

In this paper, the problem of the robust tracking for two-arm condenser cleaning crawler-type mobile manipulators (CCCMM) with delayed angle-velocity uncertainties is original investigated. The two-arm condenser cleaning crawler-type mobile manipulators are composed of a crawler-type mobile platform and two-arm industrial manipulators.The uncertainty is nonlinear time-varying and does not require a matching condition. A wavelet transform and probabilistic neural network (WTPNN) system is used to approximate an unknown controlled system from the strategic manipulation of the model following tracking errors. Based on the Lyapunov method and the linear matrix inequality (LMI) approach, several sufficient conditions, which guarantee the state variables of the closed loop system to converge, globally, uniformly and exponentially, to a ball in the state space with any pre-specified convergence rate, are derived. Experiment results are given to illustrate the superior control performance of the proposed intelligent control method.

The idea of creating a set of neural network models consists in that, due to the widespread use of distributed power generation and an increasingly frequent use of low- and medium-capacity gas-turbine power plants, it is required to ensure a stable and reliable production of electricity at required quality levels. The solution proposed to solve this task is to develop intelligent control systems and new control algorithms for gas-turbine power plants, considering the behavior of the whole structure of the electrical system on the basis of the model-oriented approach with the help of a set of neural network models. This approach opens a broad range of possibilities for studying and taking into account, first of all, the whole diversity of performance situations and design layouts of the power system and, second, the whole diversity of advanced techniques of the theory of automatic control for running individual energy modules in performance situations emerging in the energy system.

Traditional neural network is a nonlinear dynamic system formed by a large number of neurons connected to each other, it can be widely used in many research fields, such as data mining, system identification and intelligent control. The neural networks can not only deal with data problems through the unique thinking of the human brain, but can also solve the multi input and multi output problem which is difficult to be completed by the traditional computer. In this paper, in the sense of an equidistance dissection of two-dimensional input space, some new connection weights and thresholds are determined by the interpolations and arithmetic mean values of some data pairs at the adjacent intersecting points, respectively, and a new forward interpolation neural network is constructed by a step path of two-dimensional dissection. Secondly, it is proved that the interpolation neural network can approximate to a continuous function by using characteristic properties of a Sigmoidal activation function. Finally, the approximation performance of the constructed network is tested in the light of the t- hypothesis test method in statistical inference, and the approximation errors of the interpolation network are compared with that of another polynomial neural network.

In this paper, an adaptive intelligent control scheme is presented to investigate the problem of adaptive tracking control for a class of nonstrict-feedback nonlinear systems with constrained states and unmodeled dynamics. By approximating the unknown nonlinear uncertainties, utilizing Barrier Lyapunov functions (BLFs), and designing a dynamic signal to deal with the constrained states and the unmodeled dynamics, respectively, an adaptive neural network (NN) controller is developed in the frame of the backstepping design. In order to simplify the design process, the nonstrict-feedback form is treated by using the special properties of Gaussian functions. The proposed adaptive control scheme ensures that all variables involved in the closed-loop system are bounded, the corresponding state constraints are not violated. Meanwhile, the tracking error converges to a small neighborhood of the origin. In the end, the proposed intelligent design algorithm is applied to one-link manipulator to demonstrate the effectiveness of the obtained method.

The active control of a suspension system is meant to provide an isolated behavior of the system spring-mass (for example, increased comfort and performance). During this article, we are going to explain the importance of developing an intelligent control approach for active truck suspensions based on the artificial neural network. From where the main objective of this article is to obtain a mathematical model for active suspension systems then build a hydraulic model for active suspension control for trucks using an artificial neural network. In this article, a corresponding artificial neural network nonlinear active suspension controller has been designed and optimized for approximate road profiles, using simulation according to International Organization for Standardization 2631-5 and International Organization for Standardization 8608 standardizations. The model developed with MATLAB Toolbox, estimated and validated from data collected during tests carried out with a truck in other research work. To model the system, the laws of physics are used to describe the system and experimental data or information supplied about the system to determine the parameters of the system. The statement of the problem of this research is to develop a robust artificial neural network controller for the nonlinear active suspension system of the heavy truck that can improve the performances and its verifications using graphical and simulation output. The results of the simulation show that the methodology offers excellent performance. In addition, the robustness of the artificial neural network hydraulic controller is demonstrated for a variety of road profiles that increase the capabilities of the proposed methodology and prove its effectiveness.

This paper proposes an original approach to the design of intelligent control systems for complex time-varying (nonstationary) objects in which industrial processes are repeatedly reproduced over a certain period of time. Such reproduction allows accumulating data and knowledge about control programs, conditions and results of their implementation, and the trajectories of controlled variables. The motion of a controlled object is decomposed into perturbed and unperturbed motions. The unperturbed programmed motion of the object is described using expert systems and the concept of case-based reasoning (CBR); the perturbed motion of the object, using artificial neural networks.

Railway tunnels are more damaged by electrical corrosion from local conditions than other objects, since they contain a lot of moisture and emissions accumulated between the railway track and the rails. The analysis shows that the average service life of structures inside tunnels is shorter (on average by 40–50 %) than outside tunnels. A variant of intelligent systems for predicting electrocorrosion of tunnel structures is presented, a VR-neural network is applied to it, the advantages of which are currently recognized in different areas, including in the field of intelligent control. The study is devoted to an algorithm using a neural network of the backpropagation method to develop a system for assessing the risk of electrocorrosion in the lining of a railway tunnel

Humans naturally control their surrounding space. However, that capacity has not been fully used to build better intelligent controllers, mainly because the reaction time of a person limits the number of industrial applications. In this paper, the author proposes a method to overcome the problem of reaction time for a human in the control loop. This method, called Time Scaling Control, starts by modifying the constant times of the plant's model to the point where control is comfortable for a human. Then, the controller acquires the knowledge that was expressed during the human control stage and places it in a Neural Network, which controls both scaled and original plants. Time Scaling Control highly improves the control performance compared with a PID, in this case demonstrated by the control of a direct current motor, which cannot be controlled by a human without time scaling control due to the speed of the system. RESUMEN Los humanos controlan el espacio que los rodea de manera natural. Sin embargo, esta capacidad no se ha usado completamente para construir mejores controladores inteligentes, principalmente porque el tiempo de reacción de una persona limita el número de posibles aplicaciones industriales. En este artículo se propone un método para eliminar el problema del tiempo de reacción de un humano en un lazo de control. Este método, llamado Control con Escalamiento Temporal, comienza por modificar las constantes de tiempo del modelo de la planta, hasta el punto en el que el control sea cómodo para un humano. Entonces, el controlador adquiere el conocimiento que fue expresado durante la etapa de control humano y lo ubica en una red neuronal, la cual controla tanto la planta escalizada como la planta original. El Control con Escalamiento Temporal mejora bastante el desempeño del control en comparación con un PID, demostrado en este caso por el control de un motor de corriente directa, el cual no puede ser controlado por una persona sin el uso de escalamiento temporal por la velocidad del motor.

In this article, a novel adaptive intelligent control scheme is proposed to stabilize a chaotic nonlinear system with unknown dynamics in the presence of uncertainties and external disturbances. The proposed control scheme is a combination of terminal sliding mode, adaptive, and neural controllers. Terminal sliding mode control provides appropriate finite-time stability and robustness against uncertainties and external disturbances. A neural controller is used to provide suitable control action to eliminate chattering when the sliding surface is close to zero. For this purpose, a fuzzy module with simple rules is used to change the contributions of two neural and terminal sliding mode controllers, that is, when the sliding surface has a value around zero, the neural controller will take the control of the system to reduce chattering, and when the sliding surface is far from the zero region, the terminal sliding mode controller will control the system. Moreover, to cope with the unknown dynamics of the system, an online adaptive neural network is also used to approximate the unknown dynamics of the system. This hybrid control scheme is capable to decrease the contribution of the terminal sliding mode in the convergence region of the sliding surface which leads to the elimination of chattering. The combination of the several techniques to use the advantages of all the methods makes the proposed hybrid control scheme as an effective and practical scheme. Simulation results on a chaotic plasma torch system indicate the efficiency of the proposed control scheme in chattering elimination, high convergence, and also show the superior performance compared with the existing methods in the presence of disturbance and uncertainty.

In this article, we develop attitude tracking control methods for spacecraft as rigid bodies against model uncertainties, external disturbances, subsystem faults/failures, and limited resources. A new intelligent control algorithm is proposed using approximations based on radial basis function neural networks (RBFNNs) and adopting the tunable parameter-based variable structure (TPVS) control techniques. By choosing different adaptation parameters elaborately, a series of control strategies are constructed to handle the challenging effects due to actuator faults/failures and input saturations. With the help of the Lyapunov theory, we show that our proposed methods guarantee both finite-time convergence and fault-tolerance capability of the closed-loop systems. Finally, benefits of the proposed control methods are illustrated through five numerical examples.

Killing operation is an effective measure to restore bottom-hole pressure balance after unbalanced bottom-hole pressure shut-in. In the traditional well killing operation, the opening of the hydraulic throttle valve is manually adjusted by the throttle control box, and the manual control has the problems of uncertainty and low control precision, which makes the stability control of well killing operation a difficult problem. This paper presents a feedback control model based on a large number of real-time bottom-hole data, historical data and GA-BP neural network prediction. Through the intelligent control of throttle valve opening in the process of well killing operation, the fast, accurate and stable self-feedback control of bottom-hole pressure prediction and prediction output is realized. The analysis results show that the control model predicted by GA-BP neural network can effectively adjust the throttle opening and realize the stable and effective control of bottom-hole pressure.

In this article, the design of a fault tolerant controller for an induction furnace is presented. The fault studied was that of an actuator fault in the furnace. The design is comprised of a fault detection and identification (FDI) unit as well as a reconfiguration block. The induction furnace is modeled using Radial Basis Function (RBF) neural networks so that when an actuator fault happens, the FDI block detects and estimates the magnitude of the fault that occurred. Then based on a fault hiding strategy, the reconfiguration mechanism is activated to accommodate the destructive effects of the fault in a way that from the controller’s point of view the actuator fault is hidden. Utilization of the RBF neural networks model in the system also gives training ability to the system and raises its intelligence. In order to make the system robust against uncertainties caused by the fault, a sliding mode scheme is implemented as well. Using Lyapunov’s stability approach, the stability of the proposed method is verified. Finally, the simulation result of this method on the induction furnace system validates the effectiveness of the proposed algorithm in the presence of an actuator fault.

This paper developed the basic theories and techniques for a plant machinery diagnosis robot (MDR). The workplace of MDR is usually a large-scale plant or a place with a dangerous environment. It must autonomously carry out the condition monitoring of plant machinery with sensors in order to detect machinery faults for securing the safety of the plant. In this paper, the basic function, concept and structure of the MDR are discussed. The intelligent control method by rough set, fuzzy neural network (FNN) and self-location-azimuth correction for the MDR that patrols on the inspection route are proposed. Additionally, navigation and search method by fuzzy inference using extracted fault sound signals are also proposed, by which the MDR will be navigated to the near side of the faulty machine when the fault of a machine has been detected during inspection. Finally, examples of practical navigation and fault machine searching are also shown.

Remote monitoring of wireless networks based on artificial intelligence can build sensitive anomaly recognition mechanisms, automated event analysis engines and accurate global operation and maintenance capabilities for wireless network defense. Firstly, a simple camera is used to realize real-time acquisition and wireless transmission of video signals based on software-based MPEG-4 compression coding method. The self-developed ActiveX control with video decoding function is embedded in the webpage to realize real-time dynamic display of video information in the computer browser of the monitoring terminal. Secondly, in order to realize the intelligent control of wireless network, the research based on the reverse motion degree posture planning is carried out. The DH parameter method is used to establish the linkage coordinate system of wireless network remote monitoring, and the kinematics formula is derived. The geometric analysis method is used to calculate the motion trajectory of remote monitoring, accurately locate the various angles of remote monitoring and obtain the best motion path. Based on the Pm angle control method of fuzzy neural network, the RBF neural network, fuzzy control the combination of control and Pm control, using the self-learning ability of the neural network and the fuzzy reasoning is ability of fuzzy control. The end effector was adjusted to the target position. Finally, the established mathematical model was simulated by MATLAB, and the characteristics of wireless network remote monitoring were verified.

This letter presents a new intelligent control scheme for the accurate trajectory tracking of flexible link manipulators. The proposed approach is mainly based on a sliding mode controller for underactuated systems with an embedded artificial neural network to deal with modeling inaccuracies. The adopted neural network only needs a single input and one hidden layer, which drastically reduces the computational complexity of the control law and allows its implementation in low-power microcontrollers. Online learning, rather than supervised offline training, is chosen to allow the weights of the neural network to be adjusted in real time during the tracking. Therefore, the resulting controller is able to cope with the underactuating issues and to adapt itself by learning from experience, which grants the capacity to deal with plant dynamics properly. The boundedness and convergence properties of the tracking error are proved by evoking Barbalat's lemma in a Lyapunov-like stability analysis. Experimental results obtained with a small single-link flexible manipulator show the efficacy of the proposed control scheme, even in the presence of a high level of uncertainty and noisy signals., Comment: 4 pages, 5 figures

Classical control theory is widely used to regulate Simple Input, Simple Output (SISO) systems. However, when there are multiple inputs and multiple outputs, the complexity to control the systems increases significantly. Intelligent control theories such as artificial neural networks have proven to be an efficient alternative to implement complex systems. In addition, it is not necessary to obtain a mathematical model of the system. This paper shows the application of artificial neural networks as control elements in MIMO systems (Multiple Input, Multiple Outputs). The performance of the controller is shown in a simulation of a combined system, in which two first-order plants are interconnected. The experiments were carried out in Matlab software.

PID intelligent aeration control based on neural network is proposed to solve the problems of nonlinearity and large time delay in the aeration process of sewage treatment, and the system parameters cannot be accurately obtained. Firstly, the output of the controller is the nonlinear combination of the system output error signal e (k), the integral of E (k) and the differential of E (k), which can effectively approximate the nonlinearity existing in the aeration process and make the control effect of the system better; Secondly, due to the learning ability of BP neural network, when the parameters of aeration process change, the control system can correct the system parameters and avoid the problem that PID controller cannot deal with the time-varying parameters. Finally, the simulation study based on the sewage treatment process shows the effectiveness of the controller.

Plants need a specific environment to grow and reproduce in fine fettle. Nevertheless, climatic conditions are not stable and can impact their well-being and, consequently, harvest quality. Thus, greenhouse cultivation is one of the suitable agricultural techniques for creating and controlling the inside microclimate to be adequate for plant growth. The relevance of greenhouse control is widely recognized. The prediction of greenhouse variables using artificial intelligence methods is of great interest for intelligent control and the potential reduction in energetic and financial losses. However, the studies carried out in this context are still more or less limited and several machine learning methods have not been sufficiently exploited. The aim of this study is to predict the air conditioning electrical consumption and photovoltaic module electrical production at the smart Agro-Manufacturing Laboratory (SamLab) greenhouse, located in Albenga, north-western Italy. Different supervised machine learning methods were compared, namely, Artificial Neural Networks (ANNs), Gaussian Process Regression (GPR), Support Vector Machine (SVM) and Boosting trees. We evaluated the performance of the models based on three statistical indicators: the coefficient of correlation (R), the normalized root mean square error (nRMSE) and the normalized mean absolute error (nMAE). The results show good agreement between the measured and predicted values for all models, with a correlation coefficient R > 0.9, considering the validation set. The good performance of the models affirms the importance of this approach and that it can be used to further improve greenhouse efficiency through its intelligent control.

The collective motion of biological species has robust and flexible characteristics. Since the individual of the biological group interacts with other neighbors asymmetrically, which means the pairwise interaction presents asymmetrical characteristics during the collective motion, building the model of the pairwise interaction of the individual is still full of challenges. Based on deep learning (DL) technology, experimental data of the collective motion on Hemigrammus rhodostomus fish are analyzed to build an individual interaction model with multi-parameter input. First, a Deep Neural Network (DNN) structure for pairwise interaction is designed. Then, the interaction model is obtained by means of DNN proper training. We propose a novel key neighbor selection strategy, which is called the Largest Visual Pressure Selection (LVPS) method, to deal with multi-neighbor interaction. Based on the information of the key neighbor identified by LVPS, the individual uses the properly trained DNN model for the pairwise interaction. Compared with other key neighbor selection strategies, the statistical properties of the collective motion simulated by our proposed DNN model are more consistent with those of fish experiments. The simulation shows that our proposed method can extend to large-scale group collective motion for aggregation control. Thereby, the individual can take advantage of quite limited local information to collaboratively achieve large-scale collective motion. Finally, we demonstrate swarm robotics collective motion in an experimental platform. The proposed control method is simple to use, applicable for different scales, and fast for calculation. Thus, it has broad application prospects in the fields of multi-robotics control, intelligent transportation systems, saturated cluster attacks, and multi-agent logistics, among other fields.

With the advent of neural networks, fuzzy logic, genetic algorithms, and other soft computing methodologies, many researchers have demonstrated successful designs for intelligent control strategies that appear to outperform traditional linear feedback controls; however, industry has mostly ignored the new technology due to the lack of stability guarantees (required in most formal engineering risk-management processes). In this paper, we offer a Lyapunov-stability framework where one can place any arbitrary computational algorithm and still get guarantees of uniformly ultimately bounded (UUB) signals. We would expect an intelligent algorithm to be well-designed such that the proposed framework would not come into play unless unanticipated disturbances affect the system. But even if the intelligent algorithm was poorly designed, the resulting performance (inside our framework) would just look similar to that of a typical nonlinear neural-adaptive control. In our strategy, the intelligent algorithm trains a cerebellar model articulation controller (CMAC) neural network with arbitrarily bounded weights in order to achieve good performance, while a second CMAC trains in parallel using direct-adaptive-control laws in order to provide stability (even in the case of the first CMAC weights reaching their imposed bound). We test the strategy using a previously proposed ad hoc CMAC weight smoothing strategy, serving as the intelligent algorithm, with simulations and experiment controlling a two-link flexible-joint robot.

The back-propagation neural network is the most commonly used neural network for designing intelligent control systems for highway bridges. To improve the stability of an intelligent controller for a bridge with magnetorheological dampers and to reduce energy consumption, optimal performance was achieved by rationalising the number of hidden-layer elements and using a genetic algorithm (GA). Simulation results showed that a GA can achieve better optimisation results than those of simply optimised hidden-layer elements and can reduce energy consumption effectively.

This research paper investigates an intelligent control technique stabilizing a real-time model ofthe two-wheel inverted pendulum. The TWIP model is a highly non-linear, open-loop, and unstable system which makes control a challenge. Initially,a state-feedback controller that uses the dynamical system states and control signals to construct the precise control decision is used to stabilize the system. Later, Supervised Feed-Forward Neural Networks (SFFNN) based on back propagation Levenberg-Marquardt optimization algorithm are trained by using real-time measurements of system states and motors control signals from state-feedback controller stabilization. SFFNN control the two-wheel inverted pendulum better than a state-feedback controller.

A simulation method of an intelligent contactor is presented by using a neural network to fit the proven relationship among the flux linkage, the electrical current, and the moving core displacement of a contactor in this article. First, the neural network algorithm is trained by the operational data of a contactor driven by a basic training circuit to solve the coil current. Then, a dynamic simulation program of the contactor model is constructed via combining the algorithm and dynamic differential equations. On this basis, by means of the co-simulation technology, the point-by-point closed-loop simulation between the control module and the contactor model is carried out. Accordingly, the co-simulation of an intelligent contactor based on a neural network is completed. The simulation method can avoid the complex finite-element modeling of a contactor and realize the model extraction of an arbitrary contactor. The extracted model can be combined with a drive circuit and any control strategy to perform the co-simulation, which is convenient for the flexible design of hardware control circuits and software control strategies of various intelligent contactors.

This paper proposes a real-time online optimal (RT-OPT) control method based on machine learning for a current-fed dual active bridge (CF-DAB). The basis of this control strategy is the linear quadratic optimal control, which designs the sliding surface and realizes power control based on sliding mode control (SMC). For the parameters of Q and R in the objective function of the linear quadratic regulator (LQR), a genetic algorithm is used to find the optimal value, and the optimal value is taken as the sample data. Through machine learning offline training, a neural network is obtained and run online to realize real-time online optimal control. The control method was verified by simulations in MATLAB/Simulink. The RT-OPT method achieves the expected functionality, and has better dynamic and steady-state performance than the PI controller.

In view of the problems of large overshoot and large oscillation frequency in cur?rent furnace temperature control, based on the development of intelligent control theory, expert control, fuzzy control, and neural network control in intelligent control theory are combined with proportional integral derivative (PID) control. The intelligent PID control algorithm is used to carry out numerical simulation and experimental research on these several control algorithms. The results show that the adjustment effect of the intelligent PID control algorithm is significantly better than the traditional PID control algorithm. Among them, the fuzzy self-tuning PID control algorithm and the fuzzy immune PID control algorithm are feasible in the application of furnace temperature control. The neural network PID control algorithm It also has good development and application potential.

Piezoelectric actuators (PAs) require high precision positioning for the applications of micro electrical mechanical systems, but it exhibits hysteresis nonlinearity which deteriorates positioning accuracy if no proper compensation is given. Hysteresis nonlinear modeling of PAs is a prime choice for hysteresis compensation. This paper proposes a novel intelligent positioning control algorithm based on Bouc-Wen (BW) model for the compensation of a bi-morph type piezoelectric actuator (PA) suffering rate-dependent hysteresis. A region based mixed-species swarm optimization (RMSO) algorithm is proposed for BW modeling to capture the dynamic nonlinearity of a piezoelectric actuator which exhibits rate-dependent hysteresis. Results of numerical simulations have been disclosed to illustrate the performance enhancement of RMSO over classical algorithm while they are applied to the parameter fitting problem of BW model for experimentally acquired datasets. An model based adaptive Fuzzy neural network (Fuzzy-NN) controller of PA is utilized to compensate the hysteresis for the positioning tracking control. Experimental results also illustrate the good performance of the proposed RMSO-BW based control scheme for the hysteresis compensation control of the PA.

Artificial intelligence methods are widely applied in advanced control practices such as fuzzy calculation-based intelligent control method, neural network-based predictive control method, etc. Control performances are expected to be improved such as: faster control speed, smaller steady-state error, and less repeated manual tuning workloads in harmful environments for engineers. Main works are as follows: Firstly, change ratio-based fuzzy adjusted PID (FA-PID) method is improved. The adjusted parameters of FA-PID are the multiplication result of the change ratioa at the current control cycle and the control parameters at the time of previous adjustment cycle. Secondly, prediction of back propagation neural networks-based fuzzy-PID (BPNN-F-PID) and prediction of back propagation neural networks-based FA-PID (BPNN-FA-PID) are improved, in which the adjusted control parameters are calculated according to the predicted output of control system. Thirdly, comparative simulations of all the above methods are implemented, a series of better effects are found. Effects are as follows: The improved controllers have better performance such as faster control speed, better ability of anti-interference, restrained overshoot and smaller steady-state error.

The roll motions of ships advancing in heavy seas have severe impacts on the safety of crews, vessels, and cargoes; thus, it must be damped. This study presents the design of a rudder roll damping autopilot by utilizing the dual extended Kalman filter (DEKF)–trained radial basis function neural networks (RBFNN) for the surface vessels. The autopilot system constitutes the roll reduction controller and the yaw motion controller implemented in parallel. After analyzing the advantages of the DEKF-trained RBFNN control method theoretically, the ship’s nonlinear model with environmental disturbances was employed to verify the performance of the proposed stabilization system. Different sailing scenarios were conducted to investigate the motion responses of the ship in waves. The results demonstrate that the DEKF RBFNN–based control system is efficient and practical in reducing roll motions and following the path for the ship sailing in waves only through rudder actions.

The paper considers the use of artificial neural networks in order to synthesize intelligent systems governed by a synergetic control law. It has been shown that so far all the studied objects and, therefore, control laws, have been considered linear, or have been treated to reduce them to such, thereby compromising their certain features. However, as evidenced by practice, actual objects are mostly nonlinear. Consideration of such objects with an attempt of their linearization leads to that the important characteristics of the entire process are lost. A greenhouse complex is mostly composed of such nonlinear objects of control. A greenhouse, as well as each process separately, are not exception. We have proposed basic provisions to the synergistic approach related to the systems synthesis task. The synergistic synthesis of control law has been shown for a greenhouse complex under conditions of non-controlling changes in the technological parameters and external disturbances. The applied mathematical apparatus of fuzzy logic enables the implementation of fuzzy control. It manifests itself particularly positively under conditions when the processes are difficult to analyze by using conventional quantitative methods. As well as when the acquired information about the object is substandard, inaccurate, or ambiguous. This is exactly the type of information received for analysis and its subsequent use when growing vegetables at greenhouse complexes. We have proposed an algorithm to synthesize a neuro-network controller for a greenhouse complex based on the predefined synergetic control law. The algorithm is based on the performance of the synergistic controller that simulates values for temperature and humidity from an artificial neural network following our training it. A feature of the proposed integrated approach to the synthesis of an intelligent control system for a greenhouse complex is a combination of the principle of unification of the processes of self-organization and training a neural network at a preliminary stage. Such a combination ensures further stable functioning of the system aimed to intelligently control the cultivation of vegetable produce.