Digital processing of speech signals

Energy is one of the fundamental concepts in science and engineering practice, where it is common to view dynamical systems as energy-transformation devices. This perspective is particularly useful in studying complex nonlinear systems by decomposing them into simpler subsystems that, upon interconnection, add up their energies to determine the full system's behavior. The action of a controller may also be understood in energy terms as another dynamical system. The control problem can then be recast as finding a dynamical system and an interconnection pattern such that the overall energy function takes the desired form. This energy-shaping approach is the essence of passivity-based control (PBC), a controller design technique that is very well known in mechanical systems. Our objectives in the article are threefold. First, to call attention to the fact that PBC does not rely on some particular structural properties of mechanical systems, but hinges on the more fundamental (and universal) property of energy balancing. Second, to identify the physical obstacles that hamper the use of standard PBC in applications other than mechanical systems. In particular, we show that standard PBC is stymied by the presence of unbounded energy dissipation, hence it is applicable only to systems that are stabilizable with passive controllers. Third, to revisit a PBC theory that has been developed to overcome the dissipation obstacle as well as to make the incorporation of process prior knowledge more systematic. These two important features allow us to design energy-based controllers for a wide range of physical systems.

/pdf/putting-energy-back-in-control-1b6hr2kss0.pdf

Putting energy back in control

본 논문에서는 인공 신경망의 일종인 Extreme Learning Machine의 학습 알고리즘을 기반으로 하여 노이즈에 강한 특성을 보이는 퍼지 집합 이론을 이용한 새로운 패턴 분류기를 제안 한다. 기존 인공 신경망에 비해 학습속도가 매우 빠르며, 모델의 일반화 성능이 우수하다고 알려진 Extreme Learning Machine의 학습 알고리즘을 퍼지 패턴 분류기에 적용하여 퍼지 패턴 분류기의 학습 속도와 패턴 분류 일반화 성능을 개선 한다. 제안된 퍼지 패턴 분류기의 학습 속도와 일반화 성능을 평가하기 위하여, 다양한 머신 러닝 데이터 집합을 사용한다.

Extreme Learning Machine 기반 퍼지 패턴 분류기 설계

Hand gesture recognition serves as a key for overcoming many difficulties and providing convenience for human life. The ability of machines to understand human activities and their meaning can be utilized in a vast array of applications. One specific field of interest is sign language recognition. This paper provides a thorough review of state-of-the-art techniques used in recent hand gesture and sign language recognition research. The techniques reviewed are suitably categorized into different stages: data acquisition, pre-processing, segmentation, feature extraction and classification, where the various algorithms at each stage are elaborated and their merits compared. Further, we also discuss the challenges and limitations faced by gesture recognition research in general, as well as those exclusive to sign language recognition. Overall, it is hoped that the study may provide readers with a comprehensive introduction into the field of automated gesture and sign language recognition, and further facilitate future research efforts in this area.

A review of hand gesture and sign language recognition techniques

In many real-world application, e.g., speech recognition or sleep stage classification, data are captured over the course of time, constituting a Time-Series. Time-Series often contain temporal dependencies that cause two otherwise identical points of time to belong to different classes or predict different behavior. This characteristic generally increases the difficulty of analysing them. Existing techniques often depended on hand-crafted features that were expensive to create and required expert knowledge of the field. With the advent of Deep Learning new models of unsupervised learning of features for Time-series analysis and forecast have been developed. Such new developments are the topic of this paper: a review of the main Deep Learning techniques is presented, and some applications on Time-Series analysis are summaried. The results make it clear that Deep Learning has a lot to contribute to the field.

Deep Learning for Time-Series Analysis

In this paper we present an automatic visual-based sign language translation system. Our proposed automatic sign-language translator provides a real-time English translation of the Malaysia SL. To date, there have been studies on sign language recognition based on visual approach (video camera). However, the emphasis on these works is limited to a small lexicon of sign language or solely focuses on fingerspelling, which takes different approaches respectively. In practical sense, fingerspelling is used if a word cannot be expressed via sign language. Our sign language translator can recognise both fingerspelling and sign gestures that involve static and motion signs. Trained neural networks are used to identify the signs to translate into English.

Real-Time Malaysian Sign Language Translation using Colour Segmentation and Neural Network

Nonlinear identification of a small scale unmanned helicopter using optimized NARX network with multiobjective differential evolution

Unlike general gestures, Sign Languages (SLs) are highly structured so that it provides an appealing test bed for understanding more general principles for hand shape, location and motion trajectory. Hand posture shape in other words static gestures detection and recognition is crucial in SLs and plays an important role within the duration of the motion trajectory. Vision-based hand shape recognition can be accomplished using three approaches 3D hand modelling, appearance-based methods and hand shape analysis. In this survey paper, we show that extracting features from hand shape is so essential during recognition stage for applications such as SL translators.

Vision-based hand posture detection and recognition for Sign Language — A study

The development and improvements in wind energy conversion systems (WECSs) are intensively focused these days because of its environment friendly nature. One of the attractive development is the maximum power extraction (MPE) subject to variations in wind speed. This paper has addressed the MPE in the presence of wind speed and parametric variation. This objective is met by designing a generalized global sliding mode control (GGSMC) for the tracking of wind turbine speed. The nonlinear drift terms and input channels, which generally evolves under uncertainties, are estimated using feed forward neural networks (FFNNs). The designed GGSMC algorithm enforced sliding mode from initial time with suppressed chattering. Therefore, the overall maximum power point tracking (MPPT) control is very robust from the start of the process which is always demanded in every practical scenario. The closed loop stability analysis, of the proposed design is rigorously presented and the simulations are carried out to authenticate the robust MPE.

/pdf/maximum-power-extraction-strategy-for-variable-speed-wind-3pm1kpx95u.pdf

Maximum Power Extraction Strategy for Variable Speed Wind Turbine System via Neuro-Adaptive Generalized Global Sliding Mode Controller

This paper describes one of possible parameter identification approach for a quadrotor. The unknown parameter of the quadrotor will be identified using state estimation method with the implementation of Unscented Kalman Filter (UKF). In the identification of state and parameter for nonlinear dynamic system, UKF has grown to be superior techniques. Two main processes highlighted in this paper are dynamic modeling of quadrotor and the implementation of UKF algorithm. The aim is to identify and estimate the needed parameters for an autonomous quadrotor. The obtained results demonstrate the performance of UKF based on the flight test applied to the quadrotor system.

Rini Akmeliawati

Papers

Real-Time Malaysian Sign Language Translation using Colour Segmentation and Neural Network

Nonlinear identification of a small scale unmanned helicopter using optimized NARX network with multiobjective differential evolution

Vision-based hand posture detection and recognition for Sign Language — A study

Maximum Power Extraction Strategy for Variable Speed Wind Turbine System via Neuro-Adaptive Generalized Global Sliding Mode Controller

Parameter identification of an autonomous quadrotor