http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Improving the Robustness of Lips Sensing with Evolutionary Video Processing
Takuya Akashi,Yuji Wakasa,Kanya Tanaka,Minoru Fukumi 대한전자공학회 2008 ITC-CSCC :International Technical Conference on Ci Vol.2008 No.7
In this paper, an effective method is proposed for robust lips sensing. Our objectives are high-speed lips tr acking and data acquisition of a talking person in natural scenes. Our approach is based on the Evolutionary Video Processing. This method has a trade-off between accuracy and a processing time. To solve this problem, we proposed automatic Search Domain Control method and implement this method in the Evolutionary Video Processing. The tracking accuracy is improved from 66.3% to 84.9%. The proposed method can recover from occlusion and tracking loss. Comparative experiments are presented to demonstrate the effectiveness and robustness of the proposed method.
Intelligent IMC-PID Control for Ultrasonic Motor
Shenglin Mu,Kanya Tanaka,Yuji Wakasa,Takuya Akashi,Yuki Nishimura,Masato Oka 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
Ultrasonic motor(USM) is an electronic motor which causes serious characteristic changes during operation and contains non-linearity caused by friction. For USM control application, proportional integral differential(PID) con-trol is a widespread control method. But inpractical process, setting three gains before hand is not an easy work. On the other hand, the internal model control type PID(IMC-PID) which requires only one parameter setting beforehand, mainly used in process control, is more effective relatively. However, in USM control, there are limitations of control perfor-mance on conventional fixed-gain type IMC-PID control since USM’s characteristic changes and non-linearity caused by friction. So, in this paper we propose an improved method of variable gain type IMC-PID control combined with neural network(NN) named “intelligent IMC-PID for USM”. In this proposed method, system identification method is adopted to over come serious characteristic changes and neural network is used to compensate non-linearity. And the effectiveness of the proposed method has been confirmed by experiments.
PID Controller Tuning Based on the Covariance Matrix Adaptation Evolution Strategy
Akira Yoshida,Shinji Kanagawa,Yuji Wakasa,Kanya Tanaka,Takuya Akashi 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
This paper deals with a PID control problem with constraints on sensitivity and complementary sensitivity functions, and proposes a PID controller tuning method based on the Covariance Matrix Adaptation Evolution Strategy(CMA-ES). Since the original CMA-ES is developed for an optimization problem without constraints, the PID control problem is transformed into an optimization problem without constraints by using apenalty method. Numerical examples are given to show the effectiveness of the proposed method.
Construction method of Lyapunov functions based on eigen analysis
Yuki Nishimura,Yuh Yamashita,Kanya Tanaka,Yuji Wakasa 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
In this paper, we show discretized Lyapunov functions for various nonlinear dynamical systems by using the method that the authors have proposed. We use Kushner’s scheme of difference approximation with directions and Alcaraz et al.’s quantization of Markov processes to approximate Lyapunov equations by linear Schr¨odinger-like equations. We construct time-invariant functions concerned with the solutions to Schro¨dinger-like equations. Then, we show the effectiveness of the method via an example.