The subspace Gaussian mixture model—A structured model for speech recognition

We describe a new approach to speech recognition, in which all Hidden Markov Model (HMM) states share the same Gaussian Mixture Model (GMM) structure with the same number of Gaussians in each state. The model is defined by vectors associated with each state with a dimension of, say, 50, together with a global mapping from this vector space to the space of parameters of the GMM. This model appears to give better results than a conventional model, and the extra structure offers many new opportunities for modeling innovations while maintaining compatibility with most standard techniques.     

A Bernoulli–Gaussian mixture model of donation likelihood and monetary value: An application to alumni segmentation in a university setting

Advances in computational power and enterprise technology, e.g., Customer Relationship Management (CRM) software and data warehouses, allow many businesses to collect a wealth of information on large numbers of consumers. This includes information on past purchasing behavior, demographic characteristics, as well as how consumers interact with the organization, e.g., in events, on the web. The ability to mine such data sets is crucial to an organization’s ability to deliver better customer service, as well as manage its resource allocation decisions. To this end, we formulate a Bernoulli–Gaussian mixture model that jointly describes the likelihood and monetary value of repeat transactions. In addition to presenting the model, we derive an instance of the Expectation–Maximization Algorithm to estimate the associated parameters, and to segment the consumer population.     

Comparative study between a new nonlinear model and common linear model for analysing laboratory simulated‐forest hyperspectral data

The spectral unmixing of mixed pixels is a key factor in remote sensing images, especially for hyperspectral imagery. A commonly used approach to spectral unmixing has been linear unmixing. However, the question of whether linear or nonlinear processes dominate spectral signatures of mixed pixels is still an unresolved matter. In this study, we put forward a new nonlinear model for inferring end‐member fractions within hyperspectral scenes. This study focuses on comparing the nonlinear model with a linear model. A detail comparative analysis of the fractions ‘sunlit crown’, ‘sunlit background’ and ‘shadow’ between the two methods was carried out through visualization, and comparing with supervised classification using a database of laboratory simulated‐forest scenes. Our results show that the nonlinear model of spectral unmixing outperforms the linear model, especially in the scenes with translucent crown on a white background. A nonlinear mixture model is needed to account for the multiple scattering between tree crowns and background.     

T–S fuzzy model adopted SLAM algorithm with linear programming based data association for mobile robots

This paper describes a Takagi–Sugeno (T–S) fuzzy model adopted solution to the simultaneous localization and mapping (SLAM) problem with two-sensor data association (TSDA) method. Nonlinear process model and observation model are formulated as pseudolinear models and rewritten with a composite model whose local models are linear according to T–S fuzzy model. Combination of these local state estimates results in global state estimate. This paper introduces an extended TSDA (ETSDA) method for the SLAM problem in mobile robot navigation based on an interior point linear programming (LP) approach. Simulation results are given to demonstrate that the ETSDA method has low computational complexity and it is more accurate than the existing single-scan joint probabilistic data association method. The above system is implemented and simulated with Matlab to claim that the proposed method yet finds a better solution to the SLAM problem than the conventional extended Kalman filter–SLAM algorithm.     

Multi-body model of a bobsleigh: comparison with experimental data

Bobsleigh is a winter sport in which teams make timed runs down narrow, twisting, banked, iced tracks in gravity-powered sleds. To win the race, it is of great importance to have a sled that is optimized from both the sliding and the aerodynamic point of view. However, also typical vehicle parameters, such as the weight distribution and the inclination of the steering axis, play an important role. In fact, being the friction coefficient between skates and ice very small (approx. 0.04), a change in the weight acting on the front skates may significantly modify the steering feedback of the driver thus highlighting his/her driving ability. Better trajectories imply less correction on the steering axis, and thus better performances. The inclination of the steering axis, instead, generates a torque similar to the self-aligning torque produced by road vehicle tires thus stabilizing the front skates and requiring less control actions to the sled driver.     

A novel minorization–maximization framework for simultaneous feature selection and clustering of high-dimensional count data

Pattern Analysis and Applications - Count data are commonly exploited in machine learning and computer vision applications; however, they often suffer from the well-known curse of dimensionality,...     

T–S fuzzy model identification based on a novel fuzzy c-regression model clustering algorithm

This paper proposes a novel approach for identification of Takagi–Sugeno (T–S) fuzzy model, which is based on a new fuzzy c-regression model (FCRM) clustering algorithm. The clustering prototype in fuzzy space partition is hyper-plane, so FCRM clustering technique is more suitable to be applied in premise parameters identification of T–S fuzzy model. A new FCRM clustering algorithm (NFCRMA) is presented, which is deduced from the fuzzy clustering objective function of FCRM with Lagrange multiplier rule, possessing integrative and concise structure. The proposed approach consists mainly of two steps: premise parameter identification and consequent parameter identification. The NFCRMA is utilized to partition the input–output data and identify the premise parameters, which can discover the real structure of the training data; on the other hand, orthogonal least square is exploited to identify the consequent parameters. Finally, some examples are given to verify the validity of the proposed modeling approach, and the results show the new approach is very efficient and of high accuracy.     

Decentralized linear–quadratic–Gaussian control of networked control systems with asymmetric information

The decentralized linear–quadratic–Gaussian (LQG) control problem for networked control systems (NCSs) with asymmetric information is investigated, where controller 1 shares its historical information with controller 2, and not vice versa. The asymmetry of the information structure leads to the coupling between controller 2 and estimator 1, and hence the classical separation principle fails. Through the assumption of linear control strategy, the coupling between controller 2 and estimator 1 (CCE) is decoupled, but the estimation gain is still coupled with the control gain. It is noted that the control gain conforms to the backward Riccati equation while estimation gain abides by the forward equation, which is computationally challenging. Applying the stochastic maximum principle, the solvability of the decentralized LQG control problem is reduced to that of corresponding forward and backward stochastic difference equations (FBSDEs). Further, necessary and sufficient conditions for the solvability of optimal control problem are presented by two Riccati equations, one of which is nonsymmetric. Moreover, a novel iterative forward method is proposed to calculate the coupled backward control gain and forward estimation gain.     

Development of a functional model for a master–slave combined manipulator for laparoscopic surgery

Minimally invasive surgery helps patients by accelerating postoperative recovery. However, its application is impeded because it is necessary for the surgeons performing such surgery to possess surgical skills of a high order. Therefore, a master-slave combined manipulator (MCM) has been proposed as a robotic tool that enhances the surgeon's skill in laparoscopic surgery. The master grip and the slave hand are combined through the manipulator body, and a surgeon can operate the tool near the patient. The slave hand is controlled electrically by the master grip and its position is directly controlled by the surgeon. A prototype model of the MCM has been developed. The functions of the MCM have been verified by basic evaluation tests and the MCM has been used in a preliminary animal experiment. This paper describes the concept, the basic performance and the validation of the MCM.     

Computation of sub–optimal linear controls for non–linear stochastic systems†

A successive approximation technique based on statistical linearization is developed which permits the design of sub–optimal linear controls for non–linear time–invariant processes perturbed by Gaussian random disturbances. The application of the technique to a chemical process example is given, and the validity of the results is confirmed by simulation.     

Advances in computer–human interaction for detecting facial expression using dual tree multi band wavelet transform and Gaussian mixture model

Neural Computing and Applications - In human communication, facial expressions play an important role, which carries enough information about human emotions. Last two decades, it becomes a very...     

Abnormality detection based on the Kullback–Leibler divergence for generalized Gaussian data

This paper is on abnormality detection, where the observed data under the normal condition is assumed to be independent and identically distributed (i.i.d.) and follow the generalized Gaussian distribution (GGD) with shape parameter greater than 1. The Kullback–Leibler divergence (KLD) between the estimated GGD of the observed data and the normal one is used as the test statistic. An analytical expression of the KLD is derived under the normal condition when the number of samples is large; then, two algorithms with constant and adaptive thresholds are proposed. Extensive simulated and industrial case studies are conducted to verify the analytical results and to show the effectiveness of the proposed algorithms.     

A unified approach to a state-space model for linear distributed systems†

A state-space model for a class of linear distributed-parameter systems has been obtained. The method is based on the eigenfunction integration approach to the system and provides a unified approach to linear distributed systems with both laterally distributed and variable boundary controls. The method is illustrated by a one-dimensional diffusion process.     

Development of a sequential robust–tolerance design model for a destructive quality characteristic

Robust design (RD) and tolerance design (TD) have received much attention from researchers and practitioners for more than two decades, and a number of methodologies for modeling and optimizing the RD and TD processes have been studied. However, there is ample room for improvement. Because most existing research considers RD and TD as separate research fields, the primary objective of this paper is to develop a sequential robust–tolerance design method to jointly determine the best factor settings and the closed-form solutions for the optimal specification limits. We then apply the proposed method to a destructive quality characteristic. Finally, a case study and sensitivity analyses are performed for verification purposes, and further studies are discussed.     

A Dantzig–Wolfe decomposition algorithm for linear economic model predictive control of dynamically decoupled subsystems

This paper presents a warm-started Dantzig–Wolfe decomposition algorithm tailored to economic model predictive control of dynamically decoupled subsystems. We formulate the constrained optimal control problem solved at each sampling instant as a linear program with state space constraints, input limits, input rate limits, and soft output limits. The objective function of the linear program is related directly to the cost of operating the subsystems, and the cost of violating the soft output constraints. Simulations for large-scale economic power dispatch problems show that the proposed algorithm is significantly faster than both state-of-the-art linear programming solvers, and a structure exploiting implementation of the alternating direction method of multipliers. It is also demonstrated that the control strategy presented in this paper can be tuned using a weighted ℓ₁-regularization term. In the presence of process and measurement noise, such a regularization term is critical for achieving a well-behaved closed-loop performance.     

ε-Entropy and Fractal Dimension of a State for a Gaussian Measure

Kolmogorov introduced the concept of -entropy to analyze information in classical continuous systems. The fractal dimension of a geometric set was introduced by Mandelbrot as a new criterion to analyze the geometric complexity of the set. The -entropy and the fractal dimension of a state in a general quantum system were introduced by one of the present authors (MO) in order to characterize chaotic properties of general states.In this paper, we show that -entropy of a state includes Kolmogorov's -entropy, and that the fractal dimension of a state describes fractal structure of Gaussian measures.     

Least-squares parameter estimation of linear systems with noisy input–output data

This paper studies the computational efficiency of the bias-eliminated least-squares (BELS) method recently proposed for estimating linear systems in the presence of input and output noises. It is found that the BELS method based on expanding the denominator polynomial of the system transfer function by two dimensions may involve some redundant computations due to its handling of an augmented system model in its estimation scheme. To improve the computational efficiency, a direct estimation scheme is proposed to identify the underlying noisy input–output system. Numerical results show that the computational cost can be considerably reduced using such a new estimation scheme.     

EPSILON: a versatile microcomputer program for the spectrophotometric data analysis of metal–ligand equilibria

We have developed the microcomputer program EPSILON which allows the computation of equilibrium constants and extinction coefficients of several interacting species from spectrophotometric single- and multi-wavelength pH-titration data. Novel features of EPSILON include: easy treatment of the data through a windowed environment, minimization of errors with respect to the read variable (absorbance), simultaneous plotting of both the spectrophotometric titration curve and the species distribution as a function of pH during refinement and estimation of standard deviations for the refined parameters. Several examples of applications are provided and discussed in order to illustrate the use of EPSILON. The results are compared with literature values and with those obtained by BEST from potentiometric data.     

H ∞ almost disturbance decoupling for a class of linear systems: a constructive approach

A new approach to H _∞ almost disturbance decoupling for a class of linear systems is proposed in this note by using a backstepping-like technique. It is shown that the H _∞ almost disturbance decoupling problem can always be solved for the class of linear systems, and one key advantage of the proposed approach is that the static state feedback control law for the H _∞ almost disturbance decoupling can be constructively obtained.