Network Weight Update Research Articles

Finite-time optimal control for a class of nonlinear systems with performance constraints via critic-only ADP: Theory and experiments

This paper addresses the optimal control problem within the framework of adaptive dynamic programming (ADP) for a class of nonlinear systems subjected to performance constraints. A new finite-time optimal control scheme is developed to stabilize the system by using the critic-only neural network ADP method. Compared with the existing ADP-based optimal control methods with uniformly ultimately bounded stability, the provided control scheme ensures that the controlled system's state and neural network weight estimation error are finite-time stable. It can ensure optimality, prescribed performance, and finite-time stability of the closed-loop control system simultaneously through an integration of ADP, the prescribed performance control technique, and Lyapunov theory. The designed adaptive neural network weight update law can relax the persisting excitation condition. The proposed control scheme is implemented on a robotic experiment platform to achieve trajectory tracking and verify its effectiveness.

Composite adaptive neural control for automatic carrier landing system with input saturation and output constraints

This paper investigates the automatic carrier landing control problem in the presence of model uncertainty, airwake disturbances, input saturation, and output constraints. Considering the performance requirements of the carrier-based aircraft, a composite adaptive neural controller is proposed based on the time-varying barrier Lyapunov function and backstepping control techniques. The radial basis function neural network is used to approximate the model uncertainty, where the neural network weight update law incorporating prediction and tracking errors further improves the convergence rate of the neural network and mitigates high-frequency oscillations. Furthermore, an adaptive disturbance compensation model is established to mitigate the adverse effects of airwake disturbances and estimation errors in the neural network. Based on the Lyapunov stability theory, it is proven that the proposed controller maintains the aircraft trajectory within the prescribed constraints and also ensures that all signals in the closed-loop control system are semiglobally uniformly ultimately bounded. Finally, comparative simulations are performed to demonstrate the effectiveness and superiority of the proposed composite adaptive neural control method.

Exploring the effects of diverse historical stock price data on the accuracy of stock price prediction models

In the recent years, a majority of Americans have moved to investing in stocks through algorithmic trading. Not only does this underline the public’s interest in the stock market, but it also shows the important role that artificial intelligence plays when it comes to predicting and trading stocks. These prediction models can benefit investors by improving their investment decisions and potentially increasing financial gains. In this research, we aimed to explore the impact of incorporating historical stock price data types, such as the opening prices, closing prices and highest prices, on the accuracy of stock price prediction models. The primary objective was to optimize the performance of our models. To carry out this research, we developed three supervised learning artificial intelligence models using Python: linear regression, neural network, and multiplicative weight update. Our models predicted the stock prices for Microsoft, Amazon, Google, Tesla, and Apple. First, the models used the opening prices for the past three days to predict the opening price on the fourth day. However, to enhance each model’s performance, we evaluated whether adding extra information, such as closing and highest prices, would be beneficial. We hypothesized that incorporating the opening, closing, and highest prices would yield the highest accuracy as it would provide the models with the most information and help them better predict patterns in the stock market. The results supported our hypothesis as the models’ average percent errors significantly decreased when they were given all three of these data types.

Adaptive-Dynamic-Programming-Based Robust Control for a Quadrotor UAV with External Disturbances and Parameter Uncertainties

Thiswork addresses the trajectory-tracking-control problem for a quadrotor unmanned aerial vehicle with external disturbances and parameter uncertainties. A novel adaptive-dynamic-programming-based robust control method is proposed to eliminate the effects of lumped uncertainties (including external disturbances and parameter uncertainties) and to ensure the approximate optimal control performance. Its novelty lies in that two radial basis function neural network observers with fixed-time convergence properties were first established to reconstruct the lumped uncertainties. Notably, they tune only the scalar parameters online and have low computational complexities. Subsequently, two actor–critic neural networks were designed to approximate the optimal cost functions and control policies for the nominal system. In this design, two new actor–critic neural network weight update laws are proposed to eliminate the persistent excitation condition. Then, two adaptive-dynamic-programming-based robust control laws were obtained by integrating the observer reconstruction information and the nominal control policies. The uniformly ultimately bounded stability of the closed-loop tracking control systems was ensured using the Lyapunov methodology. Finally, numerical results are shown to verify the effectiveness and superiority of the proposed control scheme.

Model-based neuroadaptive event-triggered tracking consensus control for nonlinear multiagent systems with input delay

In this article, a neuroadaptive event-triggered control methodology is proposed to tackle the leader-following tracking consensus issue of uncertain nonlinear multiagent systems (MASs) with input delay and external disturbances. The proposed approach incorporates radial bias function neural networks (RBF NNs) and the integral compensation technique, effectively addressing the issues of nonlinear uncertainty and input delay, respectively. For the sake of reducing the transmission of information, we adopt the strategy that the neural network (NN) weight update law is only changed at the event-triggered instant, and a novel adaptive model and related controller are constructed. Simultaneously, this paper establishes an impulsive dynamic system to analyze the stability of the closed-loop system by extending the Lyapunov theorem to continuous and reset dynamics. Sufficient conditions for achieving leader-following tracking consensus are derived without the Zeno behavior. Finally, a numerical example is presented to illustrate the validity of the proposed method.

A Cerebellum-Inspired Prediction and Correction Model for Motion Control of a Musculoskeletal Robot

It is an important issue that how to regulate the existing control models of musculoskeletal robots to improve the ability of motion learning and generalization. In this paper, based on the motion modulation function of the cerebellum, a cerebellum-inspired prediction and correction model is proposed to carry out feedforward regulation of the original controller. Firstly, drawing on the reservoir computing mechanism of the cerebellar granular layer, the cerebellum prediction model is established by using the echo state network. Incremental learning for the network is achieved using the replay method, which is able to process control signals with different distributions. The cerebellum prediction network can accurately predict the motion results of the robot under the action of the time-series control signals. Secondly, referring to the neural pathways of the cerebellum, the cerebellum correction model is constructed. The network learning rules are designed by drawing on the long-term potentiation and depression processes of cerebellar synaptic plasticity. The characteristics of the parameters in the network weight update equations are further analyzed. And the hyperparameter update rules of the correction network weights are proposed. To simulate the function of the cerebellum involved in the limb ballistic movement, the adaptive adjustment method for cerebellum correction duration is proposed. The cerebellum correction network can accurately modulate the original control signals by using the motion prediction results. In experiments, a musculoskeletal robot is used to verify the movement effects under the control of the cerebellum model. The results show that the cerebellum-inspired model can effectively improve the motion accuracy and learning efficiency of the musculoskeletal robot, and enhance the motion generalization ability and system robustness of the robot.

Composite neural learning based appointed-time safe approach control under full-state constraints

This article investigates composite learning based appointed-time control strategy for spacecraft rendezvous and docking subject to model uncertainties, input saturation, and actuator faults. A novel prescribed performance function (PPF) is designed to constraint all system states and ensure the property of controller insensitive to initial conditions. Combining the artificial potential function (APF) method with transformed error states, a six-degree-of-freedom (6-DOF) coupled relative motion model with full-state and motion constraints is proposed. To obtain higher control accuracy and address the problem of complex calculations in network weight update, a low-complexity composite learning based neural network (NN) weight update law is investigated. The appointed-time controller is designed by utilizing backstepping method, and the dynamic surface technique is used to solve the “dimension explosion” problem. The proposed controller guarantees the stability of the system states in appointed time without the information of target. Three simulation scenarios are carried out to validate the effectiveness and robustness of the proposed controller.

The Study on Void Fraction Prediction of Gas-liquid Two Phase Flow Based on Convolutional Neural Network

For the gas-liquid two phase flow in the horizontal pipeline, at the center angle the void fraction of the different liquid phases is calculated with the finite element simulation software, and then a soft measurement model of the void fraction is established. By comparing with traditional recursive augmented least squares (RELS), particle swarm optimization (PSO), and simulated annealing-based PSO, the void fraction soft measurement model is identified and calculated separately. The segmentation optimization results of PSO based on simulated annealing have higher accuracy and stability than RELS and PSO, but as the number of center angles increases, the relative accuracy and stability of the system will deteriorate. And the characteristic is not conducive to the calculation and analysis of data results. By combining the actual model, the convolutional neural network weight update algorithm is added to the LSTM, and the RNN-LSTM convolutional neural network is used to predict the void fraction of the second half the region. It improves the effect of RNN gradient problem on learning ability and improves learning ability. Through comparison, it is found that the convolutional neural network based on RNN-LSTM has a better prediction effect, improves the accuracy and stability of the system, and provides a new method for the measured void fraction of twophase flow.

Application of intelligent edge computing technology for video surveillance in human movement recognition and Taekwondo training

With the explosive growth of global data volume, the centralized cloud computing computing model cannot provide low-latency, high-efficiency video surveillance services. Therefore, a distributed computing model is proposed, which directly processes peripheral video data to reduce transmission pressure, the burden of the central cloud server and the processing delay of the video surveillance system. Combined with the federated learning algorithm, a light neural network model is proposed. The light neural network is used to generate calculation models in different scenarios, and the generated models are reasonably arranged in edge devices. With the continuous development of technology, human motion analysis has become a hot research field in recent years, and human motion recognition is developed on the basis of human motion analysis. With the development of society, the level of Chinese Taekwondo has improved compared to before. Therefore, Chinese Taekwondo classes have gradually developed, but behind this phenomenon there is a problem of uneven teaching quality. In order to help solve the teaching problems in the learning process of Taekwondo and further improve the level of Chinese Taekwondo, the author carried out specific research in conjunction with D.A. Cooper's empirical learning theory. Future work will continue to study federated learning algorithms, optimize general neural network weight update methods, further improve detection results, and introduce neural network compression acceleration algorithms to make the system meet real-time requirements. This research uses scientific and technological means as a guide to study technical actions and strategies, and apply these strategies to specific teaching experiments for testing.

Design of RBF Adaptive Sliding Mode Controller for A Supercavitating Vehicle

This paper proposes an adaptive sliding mode control strategy based on RBF (Radial Basis Function) neural network for the supercavitating vehicle system with model uncertainties and external disturbance. Aiming at the unknown items in the model, the control strategy compensates the unknown model uncertainty and external disturbance through the RBF neural network, and derives the neural network weight update strategy according to the Lyapunov stability theory which can guarantee the closed loop system asymptotic stability. The simulation results show that this control strategy can enable the supercavitating vehicle to track reference signal when there are model uncertainties and external disturbance, and ensure the convergence of the trajectory tracking error. Compared with the control input of the sliding mode control strategy without RBF neural network, the control input of the adaptive sliding mode control strategy with RBF neural network is also reduced, which further verifies the effectiveness of the RBF adaptive sliding mode control strategy proposed in this paper.

Efficient neural learning control of nonlinear dynamics with applications

The control of nonlinear dynamics is gaining increasing attention since many practical systems are with such kind of characteristics. To deal with the system uncertainty, in this paper, the efficient learning control using neural network is proposed for the nonlinear strict-feedback system. The whole scheme is with the back-stepping design, while the novel learning is proposed for the neural network weights update. To deal with the approximation error, the robust item is added. The stability of the closed-loop dynamics is analysed and the effectiveness of the design is verified through flight simulation.

Data-Driven Output-Feedback LQ Secure Control for Unknown Cyber-Physical Systems Against Sparse Actuator Attacks

This article investigates the secure control problem of the unknown cyber-physical systems (CPSs) with the sparse actuator attacks. First, a data-driven residual generator design method via a full-rank transformation matrix is given to detect the actuator attacks. Based on the detection mechanism, an approximate dynamic programming (ADP) approach within the output-feedback framework is then developed to solve the optimal secure control problem, where a model-dependent value function is presented by combining adaptive self-organizing map neural network and clustering technology. By using the iterative learning algorithm and designing an appropriate neural network weight update law, it is shown that the secure control strategy can guarantee the stability of the closed-loop system regardless of whether the attack occurs or not and mitigate the performance loss. Finally, a numerical simulation example and a DC motor system are used to verify the effectiveness of the proposed secure control method.

Research on Brain Signals via Artificial Neural Network and Swarm Intelligence Algorithms

Artificial Neural Networks (ANNs) that are the ability to learn from theirs environment in order to improve their performance are widely used in numerous applications. The Backpropagation (BP) Algorithm is one of the most popular and effective model of ANNs. However, since it uses gradient descent algorithm which attempts to minimize the error of the network by moving gradient of the error curve, easily get trapped at local minima. To avoid this problem, we proposed an ANNs and Swarm Intelligence (SI) method, where Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO) algorithms were operated for the Multilayer Perceptron Neural Network (MLPNN) weights update. Two Electroencephalogram (EEG) datasets were used to test the success of all algorithms including ABC-MLPNN, PSO-MLPNN and conventional-MLPNN. Compared to conventional-MLPNN, higher success values were obtained on each dataset with the proposed methods. Experimental results demonstrate that combined SI and MLPNN algorithm has been increased the success of BP algorithm by avoiding local minima.

Deep metric learning via subtype fuzzy clustering

Deep metric learning methods aim to transform data features from original scattered space to a discriminative subspace in an end-to-end way, and they have shown promising results on wide applications. Triplet loss functions are the most popular models to tackle deep metric learning problem as they simultaneously enhance separability between different classes and compactness of each class in the embedded subspace. Therefore, effective triplets selection is crucial to the classification performance. However, most of these methods only focus on mining hard negative pairs, which refer to the nearest sample pairs of different classes, while fail to take subtle cluster structure of each class into consideration. To take such information into the metric learning model, a novel scheme based on subtype fuzzy clustering is proposed in this paper. By defining a new clustering degree, we conduct fuzzy clustering for each class to mine classification-oriented subtype structure. The new clustering degree is inversely proportional to the pairwise distance, thus, we can choose the positive pairs of the highest clustering degrees directly based on the farthest distances within each class. This new sampling approach avoids off-line clustering step, for which the network weights update procedure has to be temporarily paused. In other words, our method builds positive pairs without explicit clustering degree computation or off-line clustering. Our method inputs the selected positive pairs and negative pairs into the standard triplet loss to achieve network feature learning. Experimental results show competitive metric learning performance on three benchmark datasets.

Robust adaptive learning approach to self-organizing maps

The Self-Organizing Map (SOM) has attracted much research in diverse fields of study, where it has been successfully applied to a wide range of artificial intelligence applications. In this paper, a new intelligent adaptive learning SOM, in contrast with the conventional SOM algorithm, is presented. The proposed SOM overcomes the disadvantages of the conventional SOM by deriving a new variable learning rate that can adaptively achieve the optimal weights and obtain the winner neurons with a shorter learning process time. To quantify the performance, the proposed SOM was compared with various unsupervised algorithms to examine how well the network weight update equation adapts and reaches the optimum weights with fewer iterations. Extensive experiments were conducted using eight different databases from the UCI and KEEL repositories. The proposed SOM algorithm was further compared with the conventional SOM, GF-SOM, PLSOM, and PLSOM2 algorithms. The proposed SOM algorithm showed superiority in terms of the convergence rate, Quantization Error (QE), Topology Error (TE) preserving map, and accuracy during the classification process.

Adaptive Critic-Based Event-Triggered Control for HVAC System

The heating, ventilation, and air conditioning system is an important component for achieving desired thermal condition in rooms or spaces in buildings, office complex, or airports. This paper proposes a real-time event-triggered adaptive critic controller for generating near optimal control actions to achieve desired temperatures. The desired temperatures may have variable or fixed values over time. The real-time controller is designed in two phases. Initially event-triggered control actions are generated by linear quadratic regulator for small period while the actor-critic network of controller is trained. Later, adaptive critic controller takes over for event-based actions. Hence, the event triggering conditions for both general linear and nonlinear discrete time systems using Lyapunov stability analysis are derived in this paper. The event-based actor-critic network weight update formulation and ultimate boundedness of parameters are also presented in this paper. The proposed approach has been validated for different and common temperature sets for four zones, where the control execution events are minimized to 20 $\%$ and 26 $\%$ , respectively.

Neural network-based event-triggered MFAC for nonlinear discrete-time processes

This paper is concerned with the event-triggered data-driven control problem for nonlinear discrete-time systems. An event-based data-driven model-free adaptive controller design algorithm together with constructing an adaptive event-trigger condition is developed. Different from the existing data-driven model-free adaptive control approach, an aperiodic neural network weight update law is introduced to estimate the controller parameters, and the event-trigger mechanism is activated only if the event-trigger error exceeds the threshold. Furthermore, by combining the equivalent-dynamic-linearization technique with the Lyapunov method, it is proved that both the closed-loop control system and the weight estimation error are ultimately bounded. Finally, two simulation examples are provided to demonstrate the effectiveness of the derived method.

Robust adaptive dynamic programming based online tracking control algorithm for real wheeled mobile robot with omni-directional vision system

This paper proposes a new method to design an online robust adaptive dynamic programming algorithm (RADPA) for a wheeled mobile robot which is equipped with an omni-directional vision system. To integrate kinematic and dynamic controllers into the unique controller, we transform the strict feedback system dynamics into tracking error dynamics. Then, we propose a control scheme which uses only one neural network rather than three proposed in the actor-critic-based control schemes for the two-player zero-sum game problem. A neural network weight update law is designed for approximating the solution of the Hamilton–Jacobi–Isaacs equation without knowing knowledge of internal system dynamics. To implement the scheme, we propose the online RADPA, in which control and disturbance laws are updated simultaneously in an iterative loop. The convergence and stability of the online RADPA are proven by Lyapunov techniques. Simulations and experiments on a wheeled mobile robot testbed are carried out to verify the effectiveness of the proposed algorithm.

Discrete Action Dependant Heuristic Dynamic Programming in Control of a Wheeled Mobile Robot

In presented paper we propose a discrete tracking control algorithm for a two-wheeled mobile robot. The control algorithm consists of discrete Adaptive Critic Design (ACD) in Action Dependant Heuristic Dynamic Programming (ADHDP) configuration, PD controller and a supervisory term, derived from the Lyapunov stability theorem and based on the variable structure systems theory. Adaptive Critic Designs are a group of algorithms that use two independent structures for estimation of optimal value function from Bellman equation and estimation of optimal control law. ADHDP algorithm consists of Actor (ASE - Associate Search Element) that estimates the optimal control law and Critic (ACE - Adaptive Critic Element) that evaluates quality of control by estimation of the optimal value function from Bellman equation. Both structures are realized in a form of Neural Networks (NN). ADHDP algorithm does not require a plant model (the wheeled mobile robot (WMR) model) for ACE or ASE neural network weights update procedure (in contrast with other ACD configurations e.g. Heuristic Dynamic Programming or Dual Heuristic Programming that use the plant model). In presented control algorithm Actor-Critic structure is supported by PD controller and the supervisory term, that guarantee stable implementation of tracking in an initial adaptive critic neural networks learning phase, and robustness in a face of disturbances. Verification of proposed control algorithm was realized on the two-wheeled mobile robot Pioneer-2DX.

Accelerated gradient learning algorithm for neural network weights update

This work proposes decomposition of square approximation algorithm for neural network weights update. Suggested improvement results in alternative method that converge in less iteration and is inherently parallel. Decomposition enables parallel execution convenient for implementation on computer grid. Improvements are reflected in accelerated learning rate which may be essential for time critical decision processes. Proposed solution is tested and verified on multilayer perceptrons neural network case study, varying a wide range of parameters, such as number of inputs/outputs, length of input/output data, number of neurons and layers. Experimental results show time savings up to 40% in multiple thread execution.