Search

An adaptive dynamic surface control (DSC) scheme is proposed for the multi-input multi-output attitude control of near-space hypersonic vehicles (NHV). The proposed control strategy can improve the control performance of NHV despite uncertainties and external disturbances. The proposed controller combines dynamic surface control and radial basis function neural network (RBFNN) and is designed to control the longitudinal dynamics of NHV. The DSC technique is used to handle the problem of “explosion of complexity” inherent to the conventional backstepping method. RBFNN is used to approximate the unknown nonlinear function, and a robustness component is introduced in the controller to cancel the influence of compound disturbance and improve robustness and adaptation of the system. Simulation results show that the proposed strategy possesses good robustness and fast response.

This paper proposes an anti-disturbance control scheme for the near space vehicle (NSV) based on terminal sliding mode (TSM) technique and disturbance observer method. To tackle the system uncertainty and the time-varying unknown external disturbance of the NSV, a disturbance observer based on TSM technique is designed which can render the disturbance estimate error convergent in finite time. Furthermore, an auxiliary design system is introduced to analyze the input saturation effect. Based on the developed disturbance observer and the auxiliary design system, an anti-disturbance attitude control scheme is developed for the NSV using the TSM technique to speed up the convergence of all signals in closed-loop system. For the closed-loop system, the stability is rigorously proved by using the Lyapunov method and we guarantee the finite time convergence of all closed-loop system signals in the presence of the integrated affection of the system uncertainty, the input saturation, and the unknown external disturbance. Simulation study results are given to show the effectiveness of the developed TSM anti-disturbance attitude control scheme using the disturbance observer and the auxiliary system for the NSV.

This paper describes the design of guaranteed transient performance based attitude control for the near space vehicle (NSV) with control input saturation using the backstepping method. To improve the robust controllability of the NSV, the parameter adaptive method is used to tackle the integrated effect of unknown time-varying disturbance and control input saturation. Based on the backstepping technique and parameter estimated outputs, a robust attitude control scheme is proposed for the NSV with input saturation. A novel robust attitude control scheme is then proposed based on a prescribed performance bound (PPB) which characterizes the convergence rate and maximum overshoot of the attitude tracking error. The closed-loop system stability under both the developed robust attitude control schemes is proved using Lyapunovs method and uniformly asymptotical convergence of all closed-loop signals is guaranteed. Finally, simulation results are given to show the effectiveness of both the proposed robust constrained attitude control schemes.

In this article, robust attitude control schemes are proposed for the near-space vehicle with the backlash-like hysteresis and the time-varying unknown disturbance using backstepping technique. In the robust attitude control design, the backlash-like hysteresis is described by a dynamic equation which is decomposed as a linear term and a nonlinear term. To handle the nonlinear term in the dynamic equation of the backlash-like hysteresis and the time-varying unknown disturbance, they are treated as a compounded disturbance and the nonlinear disturbance observer is designed to estimate them. Based on the developed nonlinear disturbance observer, the robust backstepping attitude control is firstly developed for the near-space vehicle. Then, the robust sliding mode attitude control scheme is presented for the near-space vehicle to further enhance the robust attitude control performance. The convergence of all closed-loop signals is rigorously proved via Lyapunov analysis method under the developed robust attitude control schemes. Finally, simulation results are presented to illustrate the effectiveness of the proposed attitude control approaches.

In this paper, a robust synchronization control scheme is proposed for chaotic systems in the presence of system uncertainties and unknown external disturbances. For the synchronization error system, the compound disturbance which is estimated using the disturbance observer cannot be directly measured. If the gain matrix is properly chosen, the disturbance observer can approximate the unknown compound disturbance well. And then, the constrained robust synchronization control scheme is presented for uncertain chaotic systems based on the output of disturbance observer. In the design of a robust synchronization control scheme, the effect of unknown control input constraint has been explicitly considered to guarantee the synchronization performance. Numerical simulation results are presented to illustrate the effectiveness of the proposed constrained synchronization control scheme for uncertain chaotic systems.

This paper is concerned with the problem of hypersonic vehicle (HSV) attitude control system in uncertain flight conditions. The problem can be expressed as the adaptive robust control for a class of uncertain nonlinear systems. Based on the description of the aerodynamic structure and the model of flight control system of a certain kind of HSV, the ideal nonlinear generalized predictive control (NGPC) law based on uncertain nonlinear model is raised first to optimize the receding-horizon criterion of tracking errors. Then the online support vector regression (SVR) is employed to identify the uncertain item in the ideal control law. It is the compensating part of the controller. In addition, the stability of the close-loop system is analyzed using the Lyapunov method. The developed control strategy is well-implemented in this HSV attitude control system, and the simulation results compared with both nominal NGPC and RBF neural network disturbance observer show the good robustness and disturbance attenuation ability of this strategy and demonstrate the efficiency of online SVR algorithm.

Generally, fuzzy approximation models require some human knowledge and experience. Operator's experience is involved in the mathematics of fuzzy theory as a collection of heuristic rules. The main goal of this paper is to present a new method for identifying unknown nonlinear dynamics such as Rössler system without any human knowledge. Instead of heuristic rules, the presented method uses the input-output data pairs to identify the Rössler chaotic system. The training algorithm is a modified Levenberg-Marquardt (L-M) method, which can adjust the parameters of each linear polynomial and fuzzy membership functions on line, and do not rely on experts' experience excessively. Finally, it is applied to training Rössler chaotic system fuzzy identification. Comparing this method with the standard L-M method, the convergence speed is accelerated. The simulation results demonstrate the effectiveness of the proposed method.

A sliding mode synchronization controller is presented with RBF neural network for two chaotic systems in this paper. The compound disturbance of the synchronization error system consists of nonlinear uncertainties and exterior disturbances of chaotic systems. Based on RBF neural networks, a compound disturbance observer is proposed and the update law of parameters is given to monitor the compound disturbance. The synchronization controller is given based on the output of the compound disturbance observer. The designed controller can make the synchronization error convergent to zero and overcome the disruption of the uncertainty and the exterior disturbance of the system. Finally, an example is given to demonstrate the availability of the proposed synchronization control method.

The two-degree-of-freedom (2DOF) hypersonic flutter dynamical system has strong aeroelastic nonlinearities, and it is very difficult to obtain a more precise mathematical model. By considering varying learning rate and σ-modification factor, a novel Levenberg– Marquardt (L–M) method is proposed, based on which, an online fuzzy approximation scheme for 2DOF hypersonic flutter model is established without any human knowledge. Compared with the standard L–M method, the proposed method can obtain faster converge speed and avoid parameter drift. Numerical simulations approve the advantages of the proposed scheme.

A new learning scheme using a periodic packet as the neuronal activation function is proposed for invariance extraction and recognition of handwritten digits. Simulation results show that the proposed network can extract the invariant feature effectively and improve both the convergence and the recognition rate.

In this paper, the sliding-mode lag synchronization control scheme is proposed based on the neural network to synchronize two different delayed chaotic systems. An integral delayed sliding surface is presented to design the sliding mode control. The lag synchronization controller is achieved by combining the RBF (radial basis function) neural network with sliding-mode control. Numerical simulations are presented to demonstrate the effectiveness of the proposed sliding-mode lag synchronization control scheme.

A decentralized feedback control scheme is proposed to synchronize linearly coupled identical neural networks with time-varying delay and parameter uncertainties. Sufficient condition for synchronization is developed by carefully investigating the uncertain nonlinear synchronization error dynamics in this article. A procedure for designing a decentralized synchronization controller is proposed using linear matrix inequality (LMI) technique. The designed controller can drive the synchronization error to zero and overcome disruption caused by system uncertainty and external disturbance.

In this paper, a robust adaptive control scheme is proposed for a class of uncertain MIMO time delay systems based on backstepping method with Radical basis function(RBF) neural network. The system uncertainty is approximated by RBF neural networks, and a parameter update law is presented for approximating the system uncertainty. In each step, the control scheme is derived in terms of linear matrix inequalities (LMI's). A robust adaptive controller is designed using backstepping and LMI method based on the output of the RBF neural networks. Finally, an example is given to illustrate the availability of the proposed control scheme.

The accurate localization of iris center is difficult since the outer boundary of iris is often occluded significantly by the eyelids. In order to solve this problem, an infrared light source un-coaxial with the camera is used to produce dark pupil image for pupil center estimation. Firstly, the 3D position of the center of cornea curvature, which is used as translational movement information of eyeball, is computed using two cameras and the coordinates of two cornea reflections on the cameras' imaging planes. Then, the relative displacement of pupil center from the projection of the cornea curvature center on 2D image is extracted, describing the rotational movement of the eyeball. Finally, the feature vector is mapped into coordinates of gazing point on the screen using artificial neural network. As for the eye region detection problem, two wide-view webcams are used, and adaptive boosting+active appearance model algorithm is adopted to limit the region of interest within a small area. The result of our experiment shows that the average root-mean-square error is 0.62 in horizontal direction and 1.05 in vertical direction, which demonstrates the effectiveness of our solution in eye gaze tracking.

With the continuous development of medical imaging technology, medical image processing has played an increasingly prominent role in computer-aided diagnosis and disease management. Kidney segmentation in abdominal computed tomography (CT) sequences is a key step. In this paper, combining with the contextual property of renal tissues, a new energy minimization model based on active contour and graph cuts is proposed for kidney extraction in CT sequence. According to the relationship between the shape difference in adjacent slices and corresponding layer thickness, the optimal search range of the contour evolution is calculated for graph cut optimization. The energy function, combining the geodesic active contour with Chan-Vese model, takes into account the boundary and regional information. Then, graph cut methods are used to optimize the discrete energy function and drive the active contour towards object boundaries. Thirty abdominal CT sequences are used to evaluate the accuracy and effectiveness of the proposed algorithm. The experimental results reveal that this approach can extract renal tissues in CT sequences effectively and the average Dice coefficient reaches about 93.7%.

Eye movement information is the key clue for recognizing the vision-dominated tasks, such as browsing the web, or watching a video. However, traditional wearable sensors are invasive and the vision-based eye trackers are very expensive and need time consuming calibration. Therefore, an activity recognition method based on eye movement analysis under one web camera is first proposed and the feasibility is assessed. First, an iris tracking method for the low quality image is proposed to acquire eye movement information. Then, five ten novel features are extracted from the horizontal and the vertical eye movement signals for activity recognition, and the optimal feature subset is selected. Finally, the support vector machine is used to assess the feasibility of the proposed method. Three experiments are designed for different applications: leave-one-out cross-validation, k-fold cross-validation, and validation after respective calibration. Experimental results show that their accuracies are 68.4%, 79.3% and 84.1%, respectively, which demonstrate the promise of eye based activity recognition using one web camera.

A synchronization controller is designed based on disturbance observer for two uncertain chaotic systems in this paper. The compound disturbance of the synchronization error system consists of non-linear uncertainties and exterior disturbance of two chaotic systems. The compound disturbance observer is proposed based on RBF neural networks and the parameter's update law is given for monitoring the compound disturbance. The output of the compound disturbance observer is used to design synchronization controller. The designed synchronization controller can make the synchronization error converge to zero. Finally, an example is given to demonstrate the availability of the proposed synchronization control method.

The intelligent control of video monitoring system in a fixed background scene is realized by comparing color histogram distance between reference image and objective image. The impacts on reference images caused by the variation with time and sunlight are eliminated by renewing reference images on-line. In order to reduce computational complexity and speed up the processing, a uniform compressional algorithm of real color space based on fuzzy method is proposed. The simulations show that an image can be processed in 10 ms, which meets the requirement of real-time processing.

An adaptive training and pruning algorithm for variable bit rate(VBR) video traffic prediction is proposed in this paper. By simulation and comparison, the adaptive neural network model proposed in this paper is shown to be promising and practically feasible in obtaining the best adaptive prediction of real-time VBR video traffic.

Finding an optimal network size is one of the major concerns when building a neural network. In using the local extended Kalman filter (EKF) algorithm, we propose an efficient approach that combines EKF training and pruning as a whole. In particular, the covariance matrix obtained along with the local EKF training can be utilized to indicate the importance of the network weights. As a result, the network size can be determined adaptively to keep pace with the changes in input characteristics. The effectiveness of this algorithm is demonstrated on generalized XOR logic function and handwritten digit recognition.

To avoid the poor performance of the Hopfield model for unequally distributed neuron states and to alleviate the dynamic range constraint of optical system, a clipped model with asymmetric clipping points is proposed. Apart from its easiness for optical implementation, the capacity and capablilty of noisy tolerance are improved greatly, compared with the former clipped model. In addition, a practical encoding method-beam-direction encoding method is proposed to implement the above clipped model. The preliminary experimental results are given.

The discussion is devoted to the adaptive H∞ control method based on RBF neural networks for uncertain nonlinear systems in this paper. The controller consists of an equivalent controller and an H∞ controller. The RBF neural networks are used to approximate the nonlinear functions and the approximation errors of the neural networks are used in the adaptive law to improve the performance of the systems. The H∞ controller is designed for attenuating the influence of external disturbance and neural network approximation errors. The controller can not only guarantee stability of the nonlinear systems, but also attenuate the effect of the external disturbance and neural networks approximation errors to reach performance indexes. Finally, an example validates the effectiveness of this method.