Nowadays, artificial intelligence (AI)-based voice conversion (VC) models generate high-quality and natural-sounding voices efficiently. While existing studies have focused on detecting voice spoofing attacks, this study explores a proactive approach to safeguard speaker identity from such threats. Attackers can exploit voice biometrics by using either advanced VC techniques or traditional methods. Although zero-shot VC has been applied for source speaker anonymization, it typically fails to protect the identity of the target speaker. To address this gap, we propose a speaker privacy preservation framework that combines zero-shot AutoVC with a multi-target voice frame fusion technique. This method anonymizes the source speaker while also preventing leakage of the target speaker’s identity by blending voice frames from multiple targets. The VCTK public dataset is employed to assess the performance of the proposed model. The average cosine similarity between the original and anonymized voices yielded a result closer to the inter-speaker voice cosine similarity, which is the most suitable benchmark for speaker anonymization. Additionally, speaker identification models trained on original and vocoded data yielded low identification accuracies of 27.4% and 41.2%, respectively. The mean opinion score (MOS) evaluation confirmed that the anonymized voice quality adeptly preserved linguistic content and naturalness. Overall, the proposed method effectively anonymizes both source and target speakers, offering robust protection against voiceprint spoofing.

As a key vehicle in disaster relief scenarios, the reliability and path-following capability of autonomous vehicles are crucial. We developed a tracked mobile robot with a robust chassis to adapt to varied road conditions. A machine learning-based nonlinear model predictive control (NMPC) algorithm, combining the XGBoost model with insights gained from the disturbances encountered during the mobile robot's operation, significantly reduces path-tracking errors. Experimental results demonstrate an average path-following error of 14 cm on smooth surfaces and 17 cm on surfaces with potholes and slopes, showcasing excellent performance across different road conditions. With accumulated experience in repeated path following, further performance improvements are achievable.

This paper presents a framework for hand gesture recognition based on 3-channel electromyography (EMG) sensors. In the framework, the start and the end points of meaningful gesture segments are detected automatically by checking the cross points of EMG signals and their moving average curves. Then, a classifier combining k-Nearest Neighbor (kNN) and Decision Tree algorithms is used for achieving gesture recognition. For gesture-based control application, a real-time interactive system has been built up for household appliance using 10 kinds of hand gestures as control commands. Our proposed framework facilitates intelligent and natural control in gesture-based interaction.

In the paper, we extend the well-known golden-section search (GSS) method to make an unprecedented attempt to do discrete sequence searches. The GSS method is originally used to find the extremum of a strictly unimodal continuous function. We apply it on searching the best threshold for discretizing continuous attribute data in decision tree problems. Compared to typical methods, the shortcomings relating to massive calculation requirements for searching threshold values are eliminated. Whether it is used along with information gain or Gini index as the measure indicator for data purity of decision tree, the algorithm produces good results. To verify the proposed method, data set provided by UCI database is used on Mat lab platform to carry out the simulation. Results indicate that under the same performance index, the discrete GSS method significantly lowers iteration numbers of searching threshold values and, hence, verify the feasibility of this algorithm.

This work is proposed to actualize an image recognition system with capacity of processing the location of a flying object in terms of depth, which is measured by an RGB-D sensor. And according to the depth of the object and the XY coordinates of image pixels, the system will convert the measured data to a three-dimensional spatial coordinate with respect to the frame of the RGB-D sensor. Then an algorithm is proposed to predict the flying path and the landing point of the object in real time. The algorithm can predict the direction of the flying object and its landing point according to, respectively, the first two and the first three measured data of the object position very effectively. Moreover, not only can the proposed scheme predict the landing site of the moving object, it can also be applied to objects of different shapes, colors and sizes.

In this paper, we will record the hand movements into the computer using hand held sensors and then recognize the gesture using the suggested algorithm. The main purpose of this research is to build a gesture recognition system that develops an easier and faster recognition algorithm. To build the recognition model, we combine the hierarchical hidden Markov model (HHMM) to represent gesture units. In terms of hardware, we obtain the acceleration of hand movement by using inertial measurement unit (IMU). In terms of software, we use the improved algorithm to decide which gesture the movement belongs to. In the experiment, we sampled the ten Arabic numerals as our recognition objects. When the user waves the IMU sensor, the computer obtains the acceleration values of X, Y axes. And, the recognition program will acknowledge which Arabic numeral is being drawn.

This paper introduces a fuzzy virtual reference model (FVRM) synthesis method for linear induction motor (LIM) speed sensorless tracking control. First, we represent the LIM as a Takagi-Sugeno fuzzy model. Second, we estimate the immeasurable mover speed and secondary flux by a fuzzy observer. Third, to convert the speed tracking control into a stabilization problem, we define the internal desired states for state tracking via an FVRM. Finally, by solving a set of linear matrix inequalities (LMIs), we obtain the observer gains and the control gains where exponential convergence is guaranteed. The contributions of the approach in this paper are threefold: 1) simplified approach--speed tracking problem converted into stabilization problem; 2) omit need of actual reference model--FVRM generates internal desired states; and 3) unification of controller and observer design--control objectives are formulated into an LMI problem where powerful numerical toolboxes solve controller and observer gains. Finally, experiments are carried out to verify the theoretical results and show satisfactory performance both in transient response and robustness.

This study proposes an LMI-based integral fuzzy model predictive control (MPC) for output regulation via piece wise Lyapunov function. In order to eliminate the system's bias and guarantee the zero-offset output regulation performance, firstly we take coordinate translation on equilibrium point and introduce an added integral state of output regulation error for the nonlinear plant, and then convert the resulting augmented system into a Takagi-Sugeno fuzzy model. Under the input constraint, an integrator based state feedback control law is computed by solving a convex optimization problem based on the infinite horizon predictive control scheme. Moreover, the proposed control design has following merits: i) reducing the conservatism by using piecewise Lyapunov function; ii) improving the transient performance by adding a decay rate condition; and iii) ensuring asymptotic stability for output regulation in the presence of time varying parameter. Finally, a numerical example is presented to demonstrate the effectiveness of the proposed integral fuzzy predictive controller.

In this paper, a sensorless speed controller for linear induction motor (LIM) is developed based on a fuzzy observer. First, the LIM is represented by a T-S fuzzy model. Next, the fuzzy observer is constructed to estimate the immeasurable states including the mover speed and secondary flux, where the observer gains are obtained by computationally solving a set of linear matrix inequalities. Based on the fuzzy observer, the synthesis using the virtual desired variable concept is applied to design the control law. Then, the exponential convergence for both estimation error and tracking error is concluded. This indicates that the proposed sensorless speed control possesses the feature with fast transient response and high robustness. Finally, experiments are carried out to verify the theoretical results and show satisfactory performance.

In this paper, we investigate the decision making problem which maximizes a cost function for a system with unknown dynamics. Only implicit message coming from future trend of the system can be obtained. After set up the framework of such an optimization problem, we focus on how to determine an optimal sequence of portfolio adjustments, and the purpose is to maximize a utility function at the end of some periods. At the portfolio application, it is crucial to identify the future evolution of the portfolio composition. To this end, the well-known Black-Scholes pricing formula for option market is used to modify the parametric dynamic series. This is because the option market and the spot market are closely related and are affected by each other. Many implicit messages of stocks can be obtained through examining their options. From the implied volatility and the open interest, the investors' viewpoints on the stock prices in the future can be extracted. Then, it can improve the conventional Markowitz portfolio to establish a one-period and a multi-period fuzzy decision maker. These efficient decision makers lie on the more reliable dynamic series of the portfolio composition, and can avoid overestimating and/or underestimating expected return and expected risk. Numerical case study on many scenarios also shows the proposed decision-making scheme exhibits the highest profit for asset allocation among several portfolio models.

A micro-grid power system consisting of a photovoltaic (PV) array panel, DC/DC converter, and a battery is considered in this research. This thesis proposes a T-S fuzzy method to deal with the power tracking problem of the power generating systems. Then, the stability analysis is carried out using Lyapunov direct method whereas the control problem is formulated into the feasibility of solving a set of linear matrix inequality (LMIs). To verify the performance, we focus on the experiment in hardware realization. To this end, we establish an experiment environment for the solar power system, which includes an SP75 solar module, a DC/DC buck converter, an A/D DS1103 card, and a real-time interface, dSPACE. The results show satisfactory performances for the proposed scheme.

In this paper, we investigate the adaptive output tracking control for fuzzy stochastic parametric strict-feedback systems. To deal with the output tracking problem, we convert it into a stabilization one via the concept of virtual desired variables. Then all the analysis and synthesis are carried out using LMI (linear matrix inequality) technique. Here we emphasize that, due to the specific feature of the strict-feedback systems, the virtual desired variables can be well-defined such that our stochastic adaptive tracking control can be well developed. From the numerical simulations, it is found that the proposed schemes are feasible and the performance is satisfactory.

This paper proposes a unified hybrid fuzzy model-based control scheme for uncertain nonholonomic systems. Compared with typical hybrid fuzzy control, the stability analysis is performed based on a new concept of constructing a semicommon Lyapunov function and a new definition called as exponential-like model following. This advancement provides a strict stability analysis but results in relaxed gain conditions. In detail, a unified hybrid Takagi-Sugeno fuzzy model is first introduced for representing well-known nonholonomic systems with a momentum conservation constraint or a no-slip constraint. Then, the hybrid fuzzy controller is derived to ensure robust nonlinear model following control, i.e., an asymptotic convergence with adjustable ultimate bound and arbitrary disturbance attenuation in an -gain sense. Furthermore, an iterative linear matrix inequality technique is proposed to guarantee the stability and avoid the need of a common positive-definite matrix. Finally, the applications are carried out on a hopping robot and a car-like mobile robot. Numerical simulations and experiment results show the expected performances.

This paper presents an effective linear matrix inequality (LMI)-based adaptive scheme to design the output tracking controller for parametric strict-feedback systems. In our previous paper, the concepts of virtual desired variables (VDVs) and, in turn, the so-called generalized kinematics are introduced to benefit the tracking design. All the VDVs can be determined without fail from the generalized kinematics via a newly defined recursive procedure if the systems are in strict-feedback form. The proposed adaptive tracking control is investigated for single-input and multi-input systems with unknown parameters. Different from the existent adaptive fuzzy control, our research is an exact approach, compared with the approach using universal approximation. Relative to the adaptive backstepping control, it makes the design procedure more straightforward. Moreover, we propose a new method called balance technique to deal with the infeasibility of LMIs for the specified structure of common positive definite matrixP, and adopt an overparameterization technique to make sure that the VDVs can be well-defined. From the numerical simulations, it is shown that the proposed scheme is very powerful with expected satisfactory performance.

Sensorless speed tracking control of permanent magnet synchronous motor (PMSM) is presented in this paper. The T-S fuzzy model is used to approximate the nonlinear system of PMSM. Then, the states of the system can be estimated by a fuzzy observer. While designing the controller and the observer, the concept of parallel distributed compensation (PDC) is used. For converting the tracking control into a stabilization problem, a new control design, called virtual-desired-variable synthesis, is porposed in this paper. Then the practical controller is derived with some adequate assumptions. The gains of controller and observer are obtained separately by solving LMIs. The performance of the proposed sensorless algorithm is verified with the simulation and experimental results of PMSM for T-S fuzzy speed tracking.

In this paper, we propose a Takagi-Sugeno (T-S) fuzzy controller for active queue management (AQM) with explicit congestion notification (ECN) to improve the performance of Internet communications. We use the queue length and incoming flow rate to measure the degree of congestion effect on Internet to regulate the queue length at a specified level. Linear matrix inequality (LMI) techniques are employed to obtain the feedback gains. To proceed farther into the verification of proposed scheme, we make use of the network simulation (NS-2) in our experiment. Comparing with other AQM schemes, the experiment results show that the proposed AQM has better performance in terms of packet loss, throughput and the fluctuation of queue length.

Fuzzy observer-based control design is proposed to deal with the output tracking problem for nonlinear systems. For the purpose of tracking design, the new concept of virtual desired variables and, in turn the so-called generalized kinematics are introduced to simplify the design procedure. In light of this concept, the design procedure is split into two steps: i) Determine the virtual desired variables from the generalized kinematics; and ii) Determine the control gains just like solving linear matrix inequalities for stabilization problem. For immeasurable state variables, output feedback design is proposed. Here, we focus on a common feature held by many physical systems where their membership functions of fuzzy sets satisfy a Lipschitz-like property. Based on this setting, control gains and observer gains can be designed separately. Moreover, zero tracking error and estimation error are concluded. Three different types of systems, including nonlinear mass-spring systems, dc-dc converters, and induction motors are considered to demonstrate the design procedure. Their satisfactory simulation results verify the proposed approach

In this paper, we propose a T-S fuzzy controller which combines the merits of: i) the capability for dealing with nonlinear systems; ii) the powerful LMI approach to obtain control gains; iii) the high performance of integral controllers; iv) the workable rigorous proof for exponential convergence of error signals; and v) the flexibility on tuning decay rate. The output regulation problems of a basic buck converter and a zero-voltage-transition (ZVT) buck converter are used as application examples to illustrate the control performance of the proposed methodology. First, we consider a general nonlinear system which can represent the large-signal models of the converters. After introducing an added integral state of output regulation error and taking coordinate translation on an equilibrium point, the resulting augmented system is represented into a Takagi-Sugeno (T-S) fuzzy model. Then, the concept of parallel distributed compensation is applied to design the control law whereby the control gains are obtained by solving linear matrix inequalities (LMIs). An interesting result is that the obtained control law is formed only by the linear state feedback signals weighted by grade functions. In addition, the robustness analysis is carried out when uncertainty and disturbance are taken into consideration. The performance of numerical simulations and practical experiments results is satisfactory.

A new control scheme is proposed to improve the system performance for Takagi–Sugeno (T–S) fuzzy system using control grade functions tuned by neural networks. First, systematic modeling method is introduced to construct the exact T–S fuzzy model for a nonlinear control system. For the T–S fuzzy model, the system uncertainty affects only the membership functions. To cope with this problem, the grade functions, resulting from the membership functions of the control rules, are tuned by a back-propagation network. On the other hand, the feedback gains of the control rules are determined by solving a set of linear matrix inequalities (LMIs) which satisfy sufficient conditions of the closed-loop stability. As a result, both stability guarantee and better performance are concluded. The scheme is applied to a ball-and-beam system example verified by numerical simulations.

In this paper, we investigate the stability conditions for linear matrix inequality (LMI)-based fuzzy control design. Especially, we focus on the dependence of the stability upon membership functions. In general, the membership functions in the rule bases of Takagi-Sugeno (T-S) fuzzy model and controllers are the same and restricted between 0 and 1. In contrast to this setting, we obtain some new results when different membership functions are considered and their values lying outside the interval of [0,1] are allowed. Applying Lyapunov equation and a convex hull of fuzzy subsystems, we first establish a relationship between the stable interval characteristic polynomial and a set of feasible LMIs. Then Kharitonov's theorem gives an insight for the solvability of stabilization problems using LMI-based design and, this leads that the membership functions have an influence on stability. On the other hand, the LMI condition leads to the well-known results for LMI-based fuzzy control design. We further indicate that the different LMI conditions arise due to the same or different membership functions and find their own applications on adaptive fuzzy control. Finally, if the unit interval constraint is removed, an LMI condition for global stability is obtained

A new continuum model complementary to traditional cellular neural networks is introduced in this note. We consider a cellular neural field formed by infinitely many cellular neurons and modelled by an integrodifferential equation. Then, using LaSalle's invariance principle on Banach space, we show that the field quantity will asymptotically converge to an equilibrium state under the condition that all equilibria of the system are isolated. From the practical sense, the convergence indicates the essential capability of retrieving message from original raw data.

We present a semi-decentralized adaptive fuzzy control scheme for cooperative multirobot systems to achieve H(infinity) performance in motion and internal force tracking. First, we reformulate the overall system dynamics into a fully actuated system with constraints. To cope with both parametric and nonparametric uncertainties, the controller for each robot consists of two parts: 1) model-based adaptive controller; and 2) adaptive fuzzy logic controller (FLC). The model-based adaptive controller handles the nominal dynamics which results in both zero motion and internal force errors for a pure parametric uncertain system. The FLC part handles the unstructured dynamics and external disturbances. An H(infinity) tracking problem defined by a novel performance criterion is given and solved in the sequel. Hence, a robust controller satisfying the disturbance attenuation is derived being simple and singularity-free. Asymptotic convergence is obtained when the fuzzy approximation error is bounded with finite energy. Maintaining the same results, the proposed controller is further simplified for easier implementation. Finally, the numerical simulation results for two cooperative planar robots transporting an object illustrate the expected performance.

We propose a chaotic fuzzy reference tracking control with immeasurable states. First, we represent the chaotic system reference model by TS fuzzy models. Then a controller design is proposed to deal with mismatched parameters between the system matrices of the plant and reference model. An observer is then designed. For different premise variables between the plant and reference model, a robust approach is used. Since simultaneous solution to both the controller and observer gains with disturbances are not trivial, a three step method is utilized. The methodology proposed above is applied to Chua's circuit in numerical simulations and DSP-based experiments.

This paper proposes an output feedback control for fuzzy descriptor systems. Using a Takagi-Sugeno fuzzy model, we design a fuzzy representation of the original nonlinear system. This fuzzy representation consists of local linear descriptor systems. An H/sup /spl infin// performance criterion is then given to attenuate disturbances to a prescribed level. Finally, a two-stage method is utilized to solve both controller and observer parameters.

This paper presents a systematic design methodology for fuzzy observer-based secure communications of chaotic systems with guaranteed robust performance. The Takagi-Sugeno fuzzy models are given to exactly represent chaotic systems. Then, the general fuzzy model of many well-known chaotic systems is constructed with only one premise variable in fuzzy rules and the same premise variable in the system output. Based on this general model, the fuzzy observer of chaotic system is given and leads the stability condition of a linear-matrix inequality problem. When taking the fuzzy observer-based design to applications on secure communications, the robust performance is presented by simultaneously considering the effects of parameter mismatch and external disturbances. Then, the error of the recovered message is stated in an H/sup /spl infin// criterion. In addition, if the communication system is free of external disturbances, the asymptotic recovering of the message is obtained in the same framework. The main results also hold for applications on chaotic synchronization. Numerical simulations illustrate that this proposed scheme yields robust performance.

Addresses synthesis approaches for signal synchronization and secure communications of chaotic systems by using fuzzy system design methods based on linear matrix inequalities (LMIs). By introducing a fuzzy modeling methodology, many well-known continuous and discrete chaotic systems can be exactly represented by Takagi-Sugeno (T-S) fuzzy models with only one premise variable. Following the general form of fuzzy chaotic models, the structure of the response system is first proposed. Then, according to the applications of synchronization to the fuzzy models that have common bias terms or the same premise variable of drive and response systems, the driving signals are developed with four different types: fuzzy, character, crisp, and predictive driving signals. Synthesizing from the observer and controller points of view, all types of drive-response systems achieve asymptotic synchronization. For chaotic communications, the asymptotical recovering of messages is ensured by the same framework. It is found that many well-known chaotic systems can achieve their applications on asymptotical synchronization and recovering messages in secure communication by using either one type of driving signals or all. Several numerical simulations are shown with expected satisfactory performance.

This paper presents synthesis approaches for synchronization and secure communications of chaotic systems by using fuzzy model-based design methods. Many well-known continuous and discrete chaotic systems can be exactly represented by T-S fuzzy models with only one premise variable. According to the applications on synchronization and signal modulation, the general fuzzy models may have either i) common bias terms; or ii) the same premise variable and driving signal. Then we propose two types of driving signals, namely, fuzzy driving signal and crisp driving signal, to deal with the asymptotical synchronization and secure communication problems for cases i) and ii), respectively. Based on these driving signals, the solutions are found by solving LMI problems. It is worthy to note that many well-known chaotic systems, such as Duffing system, Chua's circuit. Rassler's system, Lorenz system, Henon map, and Lozi map can achieve their applications on asymptotical synchronization and recovering messages in secure communication by using either the fuzzy driving signal or the crisp driving signal. Finally, several numerical simulations are shown to verify the results.

This paper presents linear matrix inequalities (LMI) based fuzzy chaotic synchronization and communication. We propose a modulated Takagi-Sugeno (T-S) fuzzy model. The modulated T-S fuzzy model is constructed by choosing the common factor or the only one variable of nonlinear terms in chaotic systems as the premise variable of fuzzy rules and output signal. Following this model, some restricting conditions required in Tanaka et al. (1998) can be relaxed. This simplified design framework can be applied to many well-known chaotic systems. Also, for chaotic communications, this modulated T-S fuzzy model illustrates asymptotical recovering of the message.

An adaptive force control, scheme is presented that enables a single-link mechanism with joint flexibility to track a desired force trajectory. A dynamic model of the link system is derived, based on which a two-stage controller is constructed. It is shown that although all the system parameters including environment stiffness are unknown except for some of their bounds, all signals inside the closed-loop system remain uniformly bounded. Moreover, the force tracking error is driven to zero asymptotically. Simulation examples are provided to demonstrate the effectiveness of the proposed controller.>