Hyun-Sik Shim

A cooperative multi-agent system and its real time application to robot soccer

The soccer robot system consists of multi agents, with highly coordinated operation and movements so as to fulfill specific objectives, even under adverse situation. The coordination of the multi-agents is associated with a lot of supplementary work in advance. The associated issues are the position correction, prevention of communication congestion, local information sensing in addition to the need for imitating the human-like decision making. A control structure for a soccer robot is designed and several behaviors and actions for a soccer robot are proposed. Modified zone defense as a basic strategy and several special strategies for fouls are applied to SOTY and MIRO teams for MIROSOT (Micro-Robot World Cup Soccer Tournament).

Variable structure control of nonholonomic wheeled mobile robot

A variable structure controller (VSC) for a nonholonomic wheeled mobile robot is proposed for tracking desired trajectories. Although the matching condition for VSC is not satisfied due to nonholonomic property of the robot having two drive wheels, the proposed algorithm guarantees bounded tracking errors by choosing the sliding surface gain properly. Yet the robot tracks its given trajectory without error if there is no initial posture and velocity errors. Simulation results show the effectiveness of the proposed controller.

A Hybrid Control Structure for Vision Based Soccer Robot System

ABSTRACTA soccer robot system consists of multi-agents with highly coordinated operations and movements so as to fulfill specific objectives, even under adverse situations. The coordination of multi-agents is associated with lot of supplementary work in advance. The associated issues are position correction, prevention of communication congestion, local information sensing in addition to the need for imitating human-like decision making. A hybrid control structure for vision-based soccer robot system is designed and several behaviors and actions for a soccer robot system are proposed. The hybrid control scheme has hierarchical and behavioral structures. The effectiveness of the proposed structure is validated by implementing the same in a real soccer robot system. Variable zone defense as a basic strategy, formation strategies and set-position play strategies for fouls applied to SOTY II team during MiroSot are discussed.

Stability and four-posture control for nonholonomic mobile robots

In this paper, a four-posture control for nonholonomic mobile robots is proposed. Based upon posture velocity error dynamics, the empirical kinematic motion requirements of a mobile robot are proposed, and the four-posture control is designed in order to generate the robustly required moving trajectory with posture error reduction. The controller is designed with four possible moving directions of linear and angular velocities. The robustness of the controller is confirmed using Lyapunov stability theory. The proposed controller is experimentally demonstrated under high velocity and acceleration conditions with different control parameters. It is shown that the four-posture control algorithm is effective and feasible, even if an unsophisticated velocity controller is used as the deriving unit of a mobile robot.

The miniature omni-directional mobile robot OmniKity-I (OK-I)

The structure of OmniKity-I (OK-I), the miniature omni-directional mobile robot, and its kinematics and dynamics models are presented. OK-I has three DOFs in the plane. A conventional two wheeled mobile robot (2-WMR) serves as the base of OK-I for translational motion. The robot body is on top of the base. The third motor controls angular motion of the body as the base turns according to the translational motion. The circular surface connector (CSC) is devised to transfer signals from the base to the body without constraining angular motion of the body. Experimental results show proper omni-directional features of OK-I.

Asymptotic control for wheeled mobile robots with driftless constraints

Abstract In this paper, an asymptotic control with driftless constraints is proposed for wheeled mobile robots based on empirical practice. The equations of motion of a designed mobile robot are derived. In order to obtain the desired performance for high velocity and consecutive motions, the motion requirements are defined and then an asymptotic control is developed by using the kinematic equations of the wheeled mobile robot. The convergence of the asymptotic control is presented by using Lyapunov stability theory. Even if a nonideal velocity controller is employed for driving wheels, it is shown that the numerical performance of the asymptotic control is effective and robustness with the appropriate selection of control parameters.

Action selection and strategies in robot soccer systems

The robots used in MiroSot (Micro-Robot World Cup Soccer Tournament) are small in size (7.5cm/spl times/75cm/spl times/7.5cm), fully/semi autonomous and without any human operators. MiroSot involves multiple robots that need to collaborate in an adversarial environment to achieve specific objectives. They have dynamic environments as other robots intentionally affect the environment in unpredictable ways. The key aspect being the need for robots not only to control themselves, but also to track and control the ball which is a passive part of the environment. In multi-robot systems the environments dynamics can be determined by other robots in addition to the uncertainty that may be inherent in the domain. The interesting theoretical issue behind MiroSot experiments is the use of soccer as a prototype example of a complex, adaptive system. And let us inculcate sports robotship in our soccer robots.

Autonomous Micro-Robot 'Kity' for Maze Contest

A table-look-up immune network is proposed to solve problems arising from autonomous microrobots trying to achieve a given goal with limited memory and calculation capacity. The method is implemented and tested with a microrobot “Kity”, with a size of less than 1 cubic inch. It is possible to generate enough rules to make the robot achieve the goal of navigating freely in a maze with a small number of sensors. Experimental results show the efficacy of immune networks in controling robots in restricted environments. Kity demonstrated the efficiency of the associated algorithm arrived at by winning first prize at the 4th and 5th International Microrobot Maze Contests held in Nagoya, Japan, in October, 1995 and 1996.

Robust control of non-holonomic wheeled mobile robot based on evolutionary programming for optimal motion

Even if the stability of a mobile robot system is guaranteed, this does not imply that its behaviors are satisfactory, or its trajectories satisfy shortest path, minimum time, minimum energy, and other constraints. To deal with these concerns, we employ an evolutionary programming (EP) algorithm to obtain the optimal control variables that govern motion control of a mobile robot. This paper focuses on evolving the control parameters used in a robust motion controller to obtain time optimal, minimum arrival error, shortest path, and minimum energy performance. The usefulness of the proposed scheme is demonstrated by computer simulations.

Design of action level in a hybrid control structure for vision based soccer robot system

A design of the action level in a hybrid control structure for a vision-based soccer robot system is proposed. The hybrid control structure is composed of four levels namely, the role, action, behavior and execution levels. The control structure, which is a combination of hierarchical and behavioral structures, can meet the behavior and the design specifications of a soccer robot system. In order to increase the system reliability and the degree of achievement in the control structure, the action level should be designed so that the actions of each agent, which is necessary in robot soccer, are assigned properly.