Ensemble Support Vector Recurrent Neural Network for Brain Signal Detection.

Brain–computer interface (BCI) P300 speller can help severely disabled patients communicate and control with external machines or robots, so that the classification methods of P300 electroencephalogram (EEG) signal play an important role in the development of BCI system and technology. In this article, a novel support vector neural network (SVNN) is proposed and developed to obtain more accurate and effective EEG classification results. It is the first time to combine linear variational inequality based primal-dual neural network with convex quadratic programming problem based on support vector machine to solve the classification problem. It has been proved that the SVNN globally converges to the optimal solution of convex optimization problem and iterates the parameters in the form of matrix, which means that the method has global convergence and parallelism. The proposed SVNN method is used to solve the classification problem of P300 EEG signals. Experimental results on dataset IIb from BCI competition II and dataset II from BCI competition III show that the accuracy of the proposed SVNN method is 100% and 98%, respectively. Compared with most of the state-of-the-art algorithms, SVNN has the highest recognition accuracy and information transfer rate.

Redundant robot manipulators are usually applied to various complex scenarios where harmful noise especially periodic calculating noise always exists. In order to avoid the task failure affected by the periodic noise, a circadian rhythms neural network (CRNN) is applied to solve motion planning problems of redundant robot manipulators suffering from the periodic noise. First, in this article, we formulate a tracking problem of redundant manipulators as a convex quadratic programming (QP) problem. Second, the QP problem is converted into a matrix equation. Third, based on the neural dynamic design method, a CRNN model is exploited and developed to solve the matrix equation of the tracking problem. Comparative simulations between the proposed CRNN and the traditional zeroing neural network show that the CRNN has better robustness and performance on solving the end-effector tracking task. Two physical experiments are conducted to further certify the effectiveness, robustness, and practicability of the proposed CRNN.

This paper proposes bounds for the objective function of convex quadratic programming problem (QPP) in general form. The method uses just the eigenvalues of the Hessian matrix. By use of the eigenvalues of the Hessian matrix and solving two simple optimisation problems, we propose an interval which contains the optimal value of QPP. This method can be useful for complicated and large scale optimisation problems as well as for the integer quadratic programming problems and also can be used as a start interval for the other existing mathematical methods for QPP. Another application of the proposed interval is to help the decision maker in real applications to estimate the bounds of the optimal solution. Thus the method is useful from both theoretical and practical approaches. This method can also be applied to solve fractional quadratic programming problems as well as for binary and mixed integer quadratic programming problems. Sensitivity analysis for the objective function is another application of the method which will be discussed. To illustrate the method, several problems are solved.

This paper presents an innovative path tracking method of model predictive control based on multi-point preview (MPP-MPC) that the original equation of state is augmented by considering a dynamics model of the multiple preview points. The path tracking issue is formulated as an optimal control problem with dynamic disturbance, i.e., the future road curvature. Hence, an ameliorative Kalman filter is adopted to fuse the information about camera and vehicle dynamics to accurately estimate the road curvature. To ensure lateral control performance, dynamic constraints are defined by the estimated road curvature to guarantee comfort and safety. In the control domain of model predictive control (MPC), the model can be approximated as a linear time-invariant model (LTIM) to reduce the computational complexity. The MPC problem can be transformed into a standard quadratic programming (QP) problem with dynamic constraints of guaranteed comfort and safety. Finally, the complex path tracking problem can be solved by the QP problem. Through lane keeping experiments, it is shown that the tracking accuracy and steering smoothness can be significantly improved by the proposed method.

Objective. Brain–computer interface (BCI) technologies have been intensely studied to provide alternative communication tools entirely independent of neuromuscular activities. Current BCI technologies use electroencephalogram (EEG) acquisition methods that require unpleasant gel injections, impractical preparations and clean-up procedures. The next generation of BCI technologies requires practical, user-friendly, nonintrusive EEG platforms in order to facilitate the application of laboratory work in real-world settings. Approach. A capacitive electrode that does not require an electrolytic gel or direct electrode–scalp contact is a potential alternative to the conventional wet electrode in future BCI systems. We have proposed a new capacitive EEG electrode that contains a conductive polymer-sensing surface, which enhances electrode performance. This paper presents results from five subjects who exhibited visual or auditory steady-state responses according to BCI using these new capacitive electrodes. The steady-state visual evoked potential (SSVEP) spelling system and the auditory steady-state response (ASSR) binary decision system were employed. Main results. Offline tests demonstrated BCI performance high enough to be used in a BCI system (accuracy: 95.2%, ITR: 19.91 bpm for SSVEP BCI (6 s), accuracy: 82.6%, ITR: 1.48 bpm for ASSR BCI (14 s)) with the analysis time being slightly longer than that when wet electrodes were employed with the same BCI system (accuracy: 91.2%, ITR: 25.79 bpm for SSVEP BCI (4 s), accuracy: 81.3%, ITR: 1.57 bpm for ASSR BCI (12 s)). Subjects performed online BCI under the SSVEP paradigm in copy spelling mode and under the ASSR paradigm in selective attention mode with a mean information transfer rate (ITR) of 17.78 ± 2.08 and 0.7 ± 0.24 bpm, respectively. Significance. The results of these experiments demonstrate the feasibility of using our capacitive EEG electrode in BCI systems. This capacitive electrode may become a flexible and non-intrusive tool fit for various applications in the next generation of BCI technologies.

In this paper, we present a recurrent neural network for solving mixed linear complementarity problems (MLCPs) with positive semi-definite matrices. The proposed neural network is derived based on an NCP function and has a low complexity respect to the other existing models. In theoretical and numerical aspects, global convergence of the proposed neural network is proved. As an application, we show that the proposed neural network can be used to solve linear and convex quadratic programming problems. The validity and transient behavior of the proposed neural network are demonstrated by using five numerical examples.

A special class of recurrent neural network termed Zhang neural network (ZNN) depicted in the implicit dynamics has recently been introduced for online solution of time-varying convex quadratic programming (QP) problems. Global exponential convergence of such a ZNN model is achieved theoretically in an error-free situation. This paper investigates the performance analysis of the perturbed ZNN model using a special type of activation functions (namely, power-sum activation functions) when solving the time-varying QP problems. Robustness analysis and simulation results demonstrate the superior characteristics of using power-sum activation functions in the context of large ZNN-implementation errors, compared with the case of using linear activation functions. Furthermore, the application to inverse kinematic control of a redundant robot arm also verifies the feasibility and effectiveness of the ZNN model for time-varying QP problems solving.

Recent advancements in brain computer interfaces (BCI) have demonstrated control of robotic systems by mental processes alone. Together with invasive BCI, electroencephalographic (EEG) BCI represent an important direction in the development of BCI systems. In the context of EEG BCI, the processing of EEG data is the key challenge. Unfortunately, advances in that direction have been complicated by a lack of large and uniform datasets that could be used to design and evaluate different data processing approaches. In this work, we release a large set of EEG BCI data collected during the development of a slow cortical potentials-based EEG BCI. The dataset contains 60 h of EEG recordings, 13 participants, 75 recording sessions, 201 individual EEG BCI interaction session-segments, and over 60 000 examples of motor imageries in 4 interaction paradigms. The current dataset presents one of the largest EEG BCI datasets publically available to date.

Multidimensional scaling is a technique for exploratory analysis of multidimensional data. The essential part of the technique is minimization of a function with unfavorable properties like multimodality, non-differentiability, and invariability with respect to some transformations. Recently various two-level optimization algorithms for multidimensional scaling with city-block distances have been proposed exploiting piecewise quadratic structure of the least squares objective function with such distances. A problem of combinatorial optimization is tackled at the upper level, and convex quadratic programming problems are tackled at the lower level. In this paper we discuss a new reformulation of the problem where lower level quadratic programming problems seem more suited for two-level optimization.

For developing brain computer interface (BCI) applications, electroencephalography (EEG) is the most widely used measurement method due to its noninvasiveness, high temporal resolution, and portability. EEG signal contains sufficient neural information about each human task, which makes the extracting, and decoding of each task-related information is still challenging, especially to improve the existing BCI performances. In this paper, we present a comparison analysis to find the most relevant features and the most suitable classification method for decoding motor imagery for EEG-based BCI. Therefore, some signal processing and machine learning techniques have applied for features extraction and classification phases. For the decomposition of EEG signal, we used three type of features [EEG signal mean, root mean square (RMS) and Relative of band power (RBP)]. In addition, we investigated an analytical comparison between three methods of classification [Support Vector Machine (SVM), Linear Discriminant Analysis and K-Nearest Neighbors]. The methods were validated using a publicly available dataset (BCI Competition IV-III-a) to discriminate between two mental states (right and left hand movements) using 10-fold cross-validation. SVM method gave better classification accuracy of 76.4% using relative band powers as potential EEG features.

Brain-computer interface (BCI) technology bridges the direct communication between the brain and machines, unlocking new possibilities for human interaction and rehabilitation. EEG-based motor imagery (MI) plays a pivotal role in BCI, enabling the translation of thought into actionable commands for interactive and assistive technologies. However, the constrained decoding performance of brain signals poses a limitation to the broader application and development of BCI systems. In this study, we introduce a convolutional Transformer network (CTNet) designed for EEG-based MI classification. Firstly, CTNet employs a convolutional module analogous to EEGNet, dedicated to extracting local and spatial features from EEG time series. Subsequently, it incorporates a Transformer encoder module, leveraging a multi-head attention mechanism to discern the global dependencies of EEG's high-level features. Finally, a straightforward classifier module comprising fully connected layers is followed to categorize EEG signals. In subject-specific evaluations, CTNet achieved remarkable decoding accuracies of 82.52% and 88.49% on the BCI IV-2a and IV-2b datasets, respectively. Furthermore, in the challenging cross-subject assessments, CTNet achieved recognition accuracies of 58.64% on the BCI IV-2a dataset and 76.27% on the BCI IV-2b dataset. In both subject-specific and cross-subject evaluations, CTNet holds a leading position when compared to some of the state-of-the-art methods. This underscores the exceptional efficacy of our approach and its potential to set a new benchmark in EEG decoding.

This paper describes a new technique for generating convex, strictly concave and indefinite (bilinear or not) quadratic programming problems. These problems have a number of properties that make them useful for test purposes. For example, strictly concave quadratic problems with their global maximum in the interior of the feasible domain and with an exponential number of local minima with distinct function values and indefinite and jointly constrained bilinear problems with nonextreme global minima, can be generated. Unlike most existing methods our construction technique does not require the solution of any subproblems or systems of equations. In addition, the authors know of no other technique for generating jointly constrained bilinear programming problems.

We introduce two classes of methods for constrained smooth nonlinear programming that are widely used in practice and that are known under the names SQP for sequential quadratic programming and SCP for sequential convex programming. In both cases, convex subproblems are formulated, in the first case a convex quadratic programming problem, in the second case a convex and separable nonlinear program. An augmented Lagrangian merit function can be applied for stabilization and for guaranteeing convergence. The methods are outlined in a uniform way, convergence results are cited, and the results of a comparative performance evaluation are shown based on a set of 306 standard test problems. In addition a few industrial applications and case studies are listed that are obtained for the two computer codes under consideration, i.e., NLPQLP and SCPIP.

A classification algorithm of an SSVEP brain-Computer interface based on CCA fusion wavelet coefficients

Brain-Computer Interfaces (BCI) is a one kind of communication system that enables control of devices or communication with others only through brain signal activities without using motor activities. The main application for BCI is to provide an alternative channel for helping disabled persons, hereafter mentioned as subjects, to communicate with the external world. This paper tries to demonstrate the performance of different machine learning algorithms based on classification accuracy. Performance has been evaluated on dataset II from BCI Competition III for the year 2004 for two subjects 'A' & 'B' and dataset IIb from BCI Competition II for the year 2003 for one subject 'C'. As a primary stage, a preprocessing was applied on the samples in order to extract the most significant features before introducing them to machine learning algorithms. The algorithms applied are Bayesian Linear Discriminant Analysis (BLDA), linear Support Vector Machine (SVM), Fisher Linear Discriminant Analysis (FLDA), Generalized Anderson's Task linear classifier (GAT), Linear Discriminant Analysis (LDA). BLDA and SVM yielded the highest accuracy for all 3 subjects. BLDA algorithm achieved classification accuracy 98%, 98% and 100%, SVM algorithm achieved 98%, 96% and 100% for subjects 'A', 'B' and 'C' respectively.