http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Study on SLAM for Indoor Blimp with Visual Markers
Tatsuya Yamada,Takehisa Yairi,Suay Halit Bener,Kazuo Machida 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
The simultaneous localization and mapping (SLAM) is an essential capability for mobile robots traveling in unknown environments where globally accurate position data is not available. In this paper, we address the SLAM problem of indoor toy blimp that has no sensors such as accelerometers and gyro except a micro camera because of the weight limits. Since it is difficult to determine the exact motion models preliminarily, we assume the motion models of the blimp. The goal of this paper is to construct a 3D map of the landmarks in environment and estimate the path taken by the indoor blimp. In this paper, we use visual markers as the landmarks, since it is difficult to detect features of the landmarks. We propose the approach to SLAM using Extended Kalman Filter (EKF) and verify the effectiveness of this approach by the experiments.
Attitude Motion Estimation Using Semisupervised Dimensionality Reduction
Masao Joko,Takehisa Yairi,Kazuo Machida 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
In this paper, we use semisupervised dimensionality reduction to estimate the attitude motion of the target. The experimental results show that we can achieve the attitude motion estimation of the target from only the images even if we can’t recognize the featue marker sin the images of the target and doesn’t know the observation model. Concretely, we use Semi supervised alignment of mani folds with graph Lap lacian to get the embedding for training stat-ically. And then, we define a out-of-sample mapping using Laplacian Eigenmap Latent Variable Model(LELVM) and estimate the attitude motion of the target dynamically using Extended Karman Filter. The experimental result shows that we can achieve the dimensionality reduction with a good accuracy, and can estimate the attitude motion dynamically for the out of sample.
A Self-modeling Autonomous Airship
Halit Bener SUAY,Takehisa YAIRI,Kazuo MACHIDA 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
A self modeling airship with a parametric autonomous controller is introduced in this paper. Modeling of airships for autonomous control is a detailed and often time-taking process if some of the parameters are unknown. Although it is possible to make reasonable assumptions in some restricted case, a model generally consists of parameters such as, weight, coordinates of the center of gravity, moments of inertia, aerodynamic force coefficients and state vector of the airship. In this paper we propose taking advantage of visual markers in order to observe and estimate the state of the airship. These observations are used as an input to a linear regression process to define the relationship between the velocity and the acceleration vector of the airship. Finally as a result of the regression process aerodynamic damping parameters and provided thrust are found.