• 한국자동차공학회논문집
• /
• 제17권5호
• /
• pp.26-30
• /
• 2009

In this paper, a robust model based fault detection for varying temperature is proposed, To develop a robust force estimation model, it needs temperature information because the force sensor's output is affected by a temperature variation. If an EPB system does not include a temperature sensor, the model has a much larger error than an EPB system with a built-in temperature sensor. Therefore, the temperature is estimated by using Ohm's law. The force model is applied with a motor current, battery voltage, operation mode, and the estimated temperature to detect a force sensor's abnormal signal fault. The residual is calculated by comparing the value of the measured force and the estimated force. Fault information is collected by using the output of the evaluated residual with the adaptive thresholds. A proposed robust model based fault detection for varying temperature was verified by HILS (Hardware in the Loop Simulation).

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1989년도 한국자동제어학술회의논문집; Seoul, Korea; 27-28 Oct. 1989
• /
• pp.177-182
• /
• 1989

In this paper a new adaptive control algorithm is derived, with the unknown manipulator and payload parameters being estimated online. In practice, we may simplify the algorithm by not explicity estimating all unknown parameters. Further, the controller must be robust to residual time-varying disturbance, such as striction or torque ripple. Also, the reference model is a simple douple integrator and the acceleration input for robot manipulator consists of a proportion and derivative controller for trajectory tracking purposes. The validity of this control is confirmed in simulation where two-link robot manipulator shows the robust performances in spite of the existing nonlinear interaction and unknown parametrictings

• Communications for Statistical Applications and Methods
• /
• 제18권6호
• /
• pp.733-739
• /
• 2011

The linear absolute shrinkage and selection operator(Lasso) method improves the low prediction accuracy and poor interpretation of the ordinary least squares(OLS) estimate through the use of $L_1$ regularization on the regression coefficients. However, the Lasso is not robust to outliers, because the Lasso method minimizes the sum of squared residual errors. Even though the least absolute deviation(LAD) estimator is an alternative to the OLS estimate, it is sensitive to leverage points. We propose a robust Lasso estimator that is not sensitive to outliers, heavy-tailed errors or leverage points.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.96.3-96
• /
• 2002

A scheme of clarifying fault feature in the residual is given for model-based fault detection and diagnosis of control systems. It is based on the residual generation using a robust filter and the noise suppresion in test statistics of the residual by multi-scale discrete wavelet transform. By clarifying the fault feature in the residual, the difficulties of existing model based approaches via adopting a threshold can be overcomed and it has advantage of taking the false alarm and missed detection into acount at the same time, which can make the fault detection and diagnosis easy and correct. To show the effectiveness of our approach, the simulation results are illustrated for a linear syste...

• International Journal of Quality Innovation
• /
• 제4권1호
• /
• pp.148-159
• /
• 2003

Variation is the archenemy of quality. To reduce or control the variation in a complex production unit, firstly we need to identify the location of the root cause of the variation. This paper discusses the detection of variability and the techniques used for reduction of variation for a production unit consisting of many processes. In the first part of this paper, the background of variability detection in production systems is introduced which is then followed by a weighted network corresponding to correlation matrix of all processes. Based on the network and clustering criterion of maximum spanning tree, a classification of all processes is derived. Furthermore, the variation of each process in a class is determined by residual analysis. In the last part, the use of methods of robust design for the processes with a larger variability is discussed.

• 음성과학
• /
• 제10권3호
• /
• pp.263-277
• /
• 2003

One of major problems in speech recognition is performance degradation due to the mismatch between the training and test environments. Recently, Stereo-based Piecewise LInear Compensation for Environments (SPLICE), which is frame-based bias removal algorithm for cepstral enhancement using stereo training data and noisy speech model as a mixture of Gaussians, was proposed and showed good performance in noisy environments. In this paper, we propose several methods to improve the conventional SPLICE. First we apply Cepstral Mean Subtraction (CMS) as a preprocessor to SPLICE, instead of applying it as a postprocessor. Secondly, to compensate residual distortion after SPLICE processing, two-stage SPLICE is proposed. Thirdly we employ phonetic information for training SPLICE model. According to experiments on the Aurora 2 database, proposed method outperformed the conventional SPLICE and we achieved a 50% decrease in word error rate over the Aurora baseline system.

• 한국음향학회지
• /
• 제18권5호
• /
• pp.35-43
• /
• 1999

본 논문에서는 크게 계산량이 증가하지 않으면서 음향 반향 제거기의 성능을 향상시킬 수 있는 잡음에 강건한 적응 알고리즘과 후처리기를 제안한다. 제안하는 적응 알고리즘은 기준 입력신호와 마이크 입력신호의 전력 합으로 적응 상수를 정규화 한다. 근단화자의 음성신호나 주변 잡음이 마이크로 입력되면 그 전력만큼 적응 상수가 작아지기 때문에 이들 주변 잡음에 의한 계수 오조정을 줄일 수 있다. 잔여 반향을 제거하기 위해서 잡음에 강건한 알고리즘과 연동하는 새로운 후처리 방법을 제안한다. 제안하는 후처리 방법은 마이크 입력신호와 추정 오차신호의 상관도를 활용한다. 잔여 반향은 마이크 입력신호의 전력으로 정규화된 상관도에 의해서 감쇠되어 제거한다. 정규화 상관도는 잔여 반향에 대해서 Wiener 필터 역할을 한다. 동시 통화시에는 추정 오차신호에는 근단화자의 음성신호가 대부분을 차지함으로 정규화 상관도는 거의 1이 되어 근단화자의 음성신호는 감쇠없이 전송된다. 반향 신호만 있을 때에는 잔여 반향은 후처리기에 의해서 대부분 감쇠되어 제거된다. 제안하는 후처리기를 이용한 음향 반향 제거기의 계산량은 NLMS 알고리즘에 비해서 크게 증가하지 않는다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.33.3-33
• /
• 2001

This paper considers a fault detection and diagnosis using estimation method in uncertain systems. In the state estimation method, we use the robust H$\infty$ FIR filtering algorithm. A novel aspect of the fault detection technique described here is that it explicitly accounts for the effects of simplified models and errors due to the linearization of nonlinear systems at an operating point.

• 전자공학회논문지B
• /
• 제32B권7호
• /
• pp.974-988
• /
• 1995

In this paper, we propose a range image segmentation algorithm using robust regression. We derive a least $\kappa$th-order square (LKS) method by generalizing the least median of squares (LMedS) method and compare it with the conventional robust regressions. The LKS is robuster against outliers than the LMedS and shows performance similar to the residual consensus (RESC). The RESC uses the predetermined number of sorted residuals, whereas the LKS uses an adaptive parameter determined by given observations rather than the a priori knowledge. Computer simulation with synthetic and real range images shows that the proposed LKS algorithm gives better performance than the conventional ones.

• 제어로봇시스템학회논문지
• /
• 제9권1호
• /
• pp.82-89
• /
• 2003

A robust motion control method is proposed fur the point-to-point position control of a XY positioning system which consists of a base cart, elastic ben and moving mass. The horizontal motion controller consists of the feedforward controller to suppress the single mode vibration of the elastic beam and the feedback controller to get the high-accuracy positioning performance of the base cart. Input preshaping vibration suppression method based on system modeling with analytic frequency equation is proposed and integrated into the robust internal-loop compensator(RIC) to increase the robustness of the whole closed-loop system The vertical motion controller is proposed based on the dual RIC structure. Through experiments, it is shown that the proposed method can stabilize the system and suppress the vibration in the presence of uncertainties and disturbances.