• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.53 no.1
• /
• pp.32-39
• /
• 2004

The color-based vision systems have been used to recognize our team robots, the opponent team robots and a ball in the robot soccer system. The color-based vision systems have the difficulty in that they are very sensitive to color variations brought by brightness changes. In this paper, a neural network trained with data obtained from various illumination conditions is used to classify colors in the modified YUV color space for the robot soccer vision system. For this, a new method to measure brightness is proposed by use of a color card. After the neural network is constructed, a look-up-table is generated to replace the neural network in order to reduce the computation time. Experimental results show that the proposed color classification method is robust under illumination variations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.4
• /
• pp.289-293
• /
• 2002

Learning Classifier System (LCS) finds a new rule set using genetic algorithm (GA). In this paper, The Zeroth Level Classifier System (ZCS) is applied to evolving the strategy of a robot soccer simulation game (SimuroSot), which is a state varying dynamical system changed over time, as GBML (Genetic Based Machine Learning) and we show the effectiveness of the proposed scheme through the simulation of robot soccer.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.496-499
• /
• 2003

Support Vector Machines (SVMs) is applied to a practical problem as one of standard tools for machine learning. The application of Reinforcement Learning (RL) and SVMs in action of mobile robot is investigated. A technique to decide the action of autonomous mobile robot in practice is explained in the paper, The proposed method is to find n basis for good action of the system under unknown environment. In multi-dimensional sensor input, the most reasonable action can be automatically decided in each state by RL. Using SVMs, not only optimal decision policy but also generalized state in unknown environment is obtained.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.471-475
• /
• 2003

This paper proposes a sensor-fusion technique where the data sets for the previous moments are properly transformed and fused into the current data sets to enable accurate measurement, such as, distance to an obstacle and location of the service robot itself. In the conventional fusion schemes, the measurement is dependent on the current data sets. As the results, more of sensors are required to measure a certain physical parameter or to improve the accuracy of the measurement. However, in this approach, instead of adding more sensors to the system, the temporal sequence of the data sets are stored and utilized for the measurement improvement. Theoretical basis is illustrated by examples and the effectiveness is proved through the simulations. Finally, the new space and time sensor fusion (STSF) scheme is applied to the control of a mobile robot in an unstructured environment as well as structured environment.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.181-184
• /
• 1998

CAM-Brain is a model to develop neural networks based in cellular automata by evolution, and finally aims at a model as and artificial brain,. In order to show the feasibility of evolutionary engineering to develop an artificial brain we have attempted to evolve a module of CAM-Brain for the problem to control a mobile robot, In this paper, we present some recent results obtained by analyzing the behaviors of the evolved neural module. Several experiments reveal a couple of problems that should be solved when CAM-Brain evolves to control a mobile robot. so that some modification of the original model is proposed to solve them. The modified CAM-Brain has evolved to behave well in a simulated environment, and a thorough analysis proves the power of evolution.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.8
• /
• pp.703-709
• /
• 2000

This paper proposes a compliance control strategy for the robot manipulators accidentally interact-ing with an unknown environment. In this proposed method each in the diagonal stiffness matrix corre-sponding to the task coordinate in a Cartesian space is adaptively adjusted during contact along the corresponding axis based on the contact force with its environment. This method can be used for both unconstrained and constrained motions without any switching mechanism which often causes undesirable instability and/or vibrational motion of the end-effector. The experimental results show the effectiveness of the proposed method by employing a two link direct drive manipulator interacting with an unknown environment.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.12
• /
• pp.1062-1067
• /
• 2005

This paper discusses a skeleton feature extraction method for ubiquitous robot system. The skeleton features are used to analyze human motion and pose estimation. In different conventional feature extraction environment, the ubiquitous robot system requires more robust feature extraction method because it has internal vibration and low image quality. The new hybrid silhouette extraction method and adaptive skeleton model are proposed to overcome this constrained environment. The skin color is used to extract more sophisticated feature points. Finally, the experimental results show the superiority of the proposed method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.3
• /
• pp.314-323
• /
• 1999

This paper introduces a methodology of active calibration of a camera pose (orientation and position) using the images of cylindrical objects that are going to be manipulated. This active calibration method is different from the passive calibration where a specific pattern needs to be located at a certain position. In the active calibration, a camera attached on the robot captures images of objects that are going to be manipulated. That is, the prespecified position and orientation data of the cylindrical object are transformed into the camera pose through the two consecutive image frames. An ellipse can be extracted from each image frame, which is defined as a circular-feature matrix. Therefore, two circular-feature matrices and motion parameters between the two ellipses are enough for the active calibration process. This active calibration scheme is very effective for the precise control of a mobile/task robot that needs to be calibrated dynamically. To verify the effectiveness of active calibration, fundamental experiments are peformed.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.3
• /
• pp.249-256
• /
• 1999

The conventional robot manipulability is decomposed into linear manipulability and angular manipulability so that they may be analysed and visualized in easy way even in the case of 3 dimensional task space with 6 variables. After the Jacobian matrix is decomposed into linear part and angular part, constraint on joint velocities is transformed into linear task velocity and angular task velocity through the decomposed Jacobian matrices. Under the assumption of redundant robot manipulators, several optimization problems which utilize the redundancy are formulated to be solved by linear programming technique or sequential quadratic programming technique. After deriving the solutions of the optimization problems, we give graphical interpretations for the solutions.

• Proceedings of the KIEE Conference
• /
• 1991.07a
• /
• pp.777-780
• /
• 1991

The practical implementation of model reference adaptive systems(MRAS) using digital computer requires the derivation of discrete-time adaptation laws. This is specially important in the case of direct driver robot and light weight manipulator where inertia changes ang gravity effects are significant. We develope a discrete-time model reference adaptive control scheme for trajectory tracking of robot manipulator. Instead of the conventional Lyapunov approach hyperstabillty theory is more appealing than the Lyapunov approach. It is better suited to discrete time systems and offers more flexibility in design by providing additional free design parameters.