The Experts below are selected from a list of 222 Experts worldwide ranked by ideXlab platform
ş Cinal - One of the best experts on this subject based on the ideXlab platform.
-
robust three Impulse Sequence input shaper design
Journal of Vibration and Control, 2007Co-Authors: S S Gurleyuk, ş CinalAbstract:This article presents an unsophisticated method for tuning the amplitudes and time locations of a three-Impulse Sequence input shaper. The method helps to solve the insufficient constraint equations directly. A wide range of shapers can be obtained using the proposed method, including Zero Vibration (ZV), Zero Vibration Derivative (ZVD) and Extra Insensitive (EI) or Specified Insensitive (SI) shapers. The Impulse amplitudes can be produced without additional derivative constraints or an initial penalty of residual vibration. It is also shown that a more robust input shaper can be obtained using the new algorithm. Experimental results from a step motor are used to support the numerical solutions.
David Mercier - One of the best experts on this subject based on the ideXlab platform.
-
ICANN - Audio-Visual Speech Recognition One Pass Learning with Spiking Neurons
Artificial Neural Networks — ICANN 2002, 2002Co-Authors: Renaud Séguier, David MercierAbstract:We present a new application in the field of Impulse neurons: audio-visual speech recognition. The features extracted from the audio (cepstral coefficients) and the video (height, width of the mouth, percentage of black and white pixels in the mouth) are sufficiently simple to consider a real time integration of the complete system. A generic preprocessing makes it possible to convert these features into an Impulse Sequence treated by the neural network which carries out the classification. The training is done in one pass: the user pronounces once all the words of the dictionary. The tests on the European M2VTS Data Base shows the interest of such a system in audio-visual speech recognition. In the presence of noise in particular, the audio-visual recognition is much better than the recognition based on the audio modality only.
-
Audio-Visual Speech Recognition One Pass Learning with Spiking Neurons
Artificial Neural Networks — ICANN 2002, 2002Co-Authors: Renaud Séguier, David MercierAbstract:We present a new application in the field of Impulse neurons: audio-visual speech recognition. The features extracted from the audio (cepstral coefficients) and the video (height, width of the mouth, percentage of black and white pixels in the mouth) are sufficiently simple to consider a real time integration of the complete system. A generic preprocessing makes it possible to convert these features into an Impulse Sequence treated by the neural network which carries out the classification. The training is done in one pass: the user pronounces once all the words of the dictionary. The tests on the European M2VTS Data Base shows the interest of such a system in audio-visual speech recognition. In the presence of noise in particular, the audio-visual recognition is much better than the recognition based on the audio modality only.
William Singhose - One of the best experts on this subject based on the ideXlab platform.
-
Reference Command Shaping Using Specified-Negative-Amplitude Input Shapers for Vibration Reduction
Journal of Dynamic Systems Measurement and Control-transactions of The Asme, 2004Co-Authors: William Singhose, Erika Biediger, Y. H. Chen, Bart MillsAbstract:Residual vibrations can be greatly reduced by using specially-shaped reference command signals. Input shaping is one such technique that reduces vibration by convolving a Sequence of Impulses with any desired reference command. Several types of useful Impulse Sequences have been developed. Most of these have contained only positively valued Impulses. However, rise time can be improved by using some negative Impulses in the Sequence. Unfortunately, the use of negative Impulses can excite unmodeled high modes. A new type of Impulse Sequence containing negative Impulses is proposed. These Sequences are designed to fill the performance gap between all-positive Impulse Sequences and the negative Sequences previously developed. A proof governing the worst case scenario provides an upper bound on high-mode excitation. The resulting class of Impulse Sequences allows the designer to make a precise trade off between rise time and vibration reduction.
-
Command generation using specified-negative-amplitude input shapers
Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251), 1999Co-Authors: William Singhose, B.w. MillsAbstract:Residual vibrations can be greatly reduced by using specially-shaped command signals. Input shaping is a technique that reduces vibration by convolving a Sequence of Impulses with any desired system command. The resulting shaped input is then used to drive the system. Many types of useful Impulses Sequences have been proposed. Most of these have contained only positively valued Impulses. However, the system rise time can be improved by using some negative Impulses. A new type of Impulse Sequence containing negative Impulses is proposed. These Sequences are designed to fill the performance gap between all-positive Sequences and the negative Sequences previously developed.
-
Time-Optimal Negative Input Shapers
Journal of Dynamic Systems Measurement and Control-transactions of The Asme, 1997Co-Authors: William Singhose, Warren Seering, Neil C SingerAbstract:Input shaping reduces residual vibration in computer controlled machines by convolving a Sequence of Impulses with a desired system command. The resulting shaped input is then used to drive the system. The Impulse Sequence has traditionally contained only positively valued Impulses. However, when the Impulses are allowed to have negative amplitudes, the rise time can be improved. Unfortunately, excitation of unmodeled high modes and overcurrenting of the actuators may accompany the improved rise time. Solutions to the problem of high-mode excitation and overcurrenting are presented. Furthermore, a simple look-up method is presented that facilitates the design of negative input shapers. The performance of negative shapers is evaluated experimentally on two systems; one driven by a piezo actuator and the other equipped with DC motors.
-
Convolved and simultaneous two-mode input shapers
IEE Proceedings - Control Theory and Applications, 1997Co-Authors: William Singhose, E. Crain, Warren SeeringAbstract:Input shaping is a method of reducing residual vibrations in computer controlled machines. Vibration is eliminated by convolving a Sequence of Impulses with a desired system command to produce a shaped input that is then used to drive the system. Techniques for designing the Impulse Sequence for two-mode systems are presented. Properties of the various techniques are compared as a function of the mode ratio. Simulations of a two-mode system demonstrate key results.
-
figure 1 input shaping by convolving desired input with an Impulse Sequence comparison of command shaping methods for reducing residual vibration
1995Co-Authors: William Singhose, Neil C Singer, Warren SeeringAbstract:This paper compares command shaping techniques for controlling residual vibration of high-performance machines. Input shaping generates vibration-reducing shaped commands through convolution of an Impulse Sequence with the desired command. Because input shaping has similarities to notch filtering, it is compared here with a variety of FIR and IIR filters. Several types of input shapers are presented and shown to be more effective than any of the conventional filters.
Jean-marc Boucher - One of the best experts on this subject based on the ideXlab platform.
-
EUSIPCO - SEM blind identification of ARMA models application to seismic data
2004Co-Authors: Benayad Nsiri, Thierry Chonavel, Jean-marc BoucherAbstract:In this paper, we address blind identification of an ARMA model convolved with an Impulse Sequence via Maximum Likelihood (ML) approach. A Stochastic Expectation Maximization (SEM) implementation of the criterion is considered. The problem of ARMA models with long Impulse response is addressed as well as the SEM initialization problem. The model estimation is performed in two steps: First, a truncated estimate of the wavelet is obtained from a SEM algorithm. Then improved wavelet estimation is achieved by fitting an ARMA model to the initial MA wavelet using the Prony algorithm. Simulation results show the significant improvement brought by this approach in situations corresponding to seismic data deconvolution.
-
SEM blind identification of ARMA models application to seismic data
2004 12th European Signal Processing Conference, 2004Co-Authors: Benayad Nsiri, Thierry Chonavel, Jean-marc BoucherAbstract:In this paper, we address blind identification of an ARMA model convolved with an Impulse Sequence via Maximum Likelihood (ML) approach. A Stochastic Expectation Maximization (SEM) implementation of the criterion is considered. The problem of ARMA models with long Impulse response is addressed as well as the SEM initialization problem. The model estimation is performed in two steps: First, a truncated estimate of the wavelet is obtained from a SEM algorithm. Then improved wavelet estimation is achieved by fitting an ARMA model to the initial MA wavelet using the Prony algorithm. Simulation results show the significant improvement brought by this approach in situations corresponding to seismic data deconvolution.
Kanakov Vladimir - One of the best experts on this subject based on the ideXlab platform.
-
EUSIPCO - The optimal detector for a wideband noise Impulse
2004Co-Authors: Panfilov Sergey, Kanakov VladimirAbstract:In this paper we discuss the problem of wideband noise Impulse Sequence detection. The structure of the optimal detector has been proposed and error detection probabilities have been estimated.
-
The optimal detector for a wideband noise Impulse
2004 12th European Signal Processing Conference, 2004Co-Authors: Panfilov Sergey, Kanakov VladimirAbstract:In this paper we discuss the problem of wideband noise Impulse Sequence detection. The structure of the optimal detector has been proposed and error detection probabilities have been estimated.
