• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.228-231
• /
• 2003

This paper presents the leg trajectory generation for the quadruped robot with genetic-fuzzy algorithm. To have the nobility even at uneven terrain, a robot is able to recognize obstacles, and generates moving path of body that can avoid obstacles. This robot should have its own avoidance algorithm against obstacles, forwarding to target without collision. During walking period, n robot recognizes obstacle from external environment with a PSD and some interface, and this obstacle information is converted into proper the body rotation angle by fuzzy inference engine. After this process, we can infer the walking direction and walking distance of body, and finally can generate the optimal Beg trajectory using genetic algorithm. All these methods are verified with PC simulation program, and implemented to SERO-V robot.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1995.10a
• /
• pp.1133-1137
• /
• 1995

This paper presents a numerical method of the global search for an optimal geometric path for a manipulator arm amid obstacles. Finite term quintic B-splines are used to describe an arbitrary point-to-point manipulator motion with fixed moving time. The coefficients of the splines span a linear vector space, a point in which uniquely represents the manipulator motion. All feasible geometric paths are searched by adjusting the seed points of the obstacle models in the penetration growth distances. In the numerical implementation using nonlinear programming, the globally optimal geometric path is obtained for a spatial 3-link(3-revolute joints) manipulator amid several hexahedral obstacles without simplifying any dynamic or geometric models.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.26 no.2
• /
• pp.47-53
• /
• 2018

In this paper, we investigate a collision avoidance algorithm for unmanned aerial vehicles using potential field based on the relative velocity of obstacles. The potential field consists of the attraction force and the repulsive force that are generated for the target and the obstacles. And the field can be classified into the attractive potential field generated by the target and the repulsive potential field generated by the obstacle, respectively. In this study, we construct an attractive potential field as a function of the distance between the UAV and the target position. On the other hand, a repulsive potential field is created by a function of distance and the relative velocity of the obstacle with respect to the UAV. The proposed potential field based collision avoidance algorithm is evaluate through simulations.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.249-251
• /
• 2005

In this paper, an algorithm of path planning and obstacle avoidance for mobile robot is proposed. We call the proposed method Random Access Sequence(RAS) method. In the proposed method, a small region is set first and numbers are assigned to its neighbors. By processing assigned numbers all regions are covered and then the path from start to destination is selected by these numbers. The RAS has an advantage of fast planning because of simple operations. This implies that new path selection may be possible within a short time and helps a robot to avoid obstacles in any direction. The algorithm can be applied to unknown environments. When moving obstacles appear, a mobile robot avoids obstacles reactively. then new path is selected by RAS.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.415-415
• /
• 2000

The functions of a mobile robot such as obstacle knowledge and collision avoidance for in-door cleaning are necessary features, as has been much studied in the field of industrial automatic guided vehicle or general mobile robot. A mobile robot, in order to avoid collision with obstacles, has to gather data with environment knowledge sensors and recognize environment and the shape of obstacles from the data. In the study, a wall-following algorithm was suggested as a autonomous moving algorithm in which a mobile robot can recognize obstacles in indoor like environment and do cleaning work in effect. The system suggested in the study is for cleaning of nuclear material dusts generated in the process of nuclear fuel manufacturing and decontamination of devices in disorder which is performed in M6 radioactive ray shield hot-cell in IMEF(Irradiated Material Examination Facility) in the Korea Atomic Energy Research Institute.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.8
• /
• pp.165-173
• /
• 1998

In this paper, a motion control algorithm is developed by using fuzzy control technique, which makes a robot arm avoid unexpected obstacles when the robot is moving from the start to a goal posture. During the motion, if there exist no obstacles the robot arm moves along the pre-defined path. But if some obstacles are recognized and close to the robot arm, a fuzzy controller is activated to adjust the path of the robot arm. To show the feasibility of the developed algorithm, numerical simulations and experiments are carried out. In the experiments, redundant planar robot arms are considered for the collision avoidance test, and it was proved that the developed algorithm gives good collision avoiding performance.

• Journal of the military operations research society of Korea
• /
• v.18 no.2
• /
• pp.140-150
• /
• 1992

In this paper, we consider the shortest path problem of a Robot car moving in a workspace which consists of some obstacles. The motion of the Robot car is considered to have initial and final directions with some restrictions in the curvature of the path. At first we consider the problem in the case of having no obstacles and we give an analytical solution. Then wre present an algorithm to find a feasible path in the case of having obstacles and a method to improve this feasible path into a minimal path. Some computational results using Graph theory and Linear programming have been included.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.943-944
• /
• 2006

In this paper, we implemented the autonomous mobile robot which can recognize and avoid obstacles, then move to its destination using a camera and ultrasonic sensors. The mobile robot can avoid both stationary obstacles with a camera and moving obstacles with ultrasonic sensors. It can find the self-location with the map-based system, that is, it attempts to localize by collecting sensor data, then updating some belief about its position with respect to a map of the environment.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.269-273
• /
• 2000

This paper proposes a algorithm for several mobile robots navigation. There are three parts in this algorithm. First part generates robots turning angle and moving distance for goal approaching, sencond part generates robots avoiding angle and avoiding distance for static obstacles or other robots and third part adjust between robots moving distance and avoiding distance. Most simulation results of this algorithm are very effective for several mobile robots traveling in unknown field.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.11
• /
• pp.1193-1199
• /
• 2014

To achieve safe and high-speed navigation of a mobile service robot, velocity of dynamic obstacles should be considered while planning the trajectory of a mobile robot. Trajectory planning schemes without considering the velocity of the dynamic obstacles may collide due to the relative velocities or dynamic constraints. However, the general planning schemes that considers the dynamic obstacle velocities requires long computational times. This paper proposes a velocity control scheme by scaling the time step of trajectory to deal with dynamic obstacle avoidance problem using the RNLVO (Robot Nonlinear Velocity Obstacles). The RNLVO computes the collision conditions on the basis of the NLVO (Nonlinear Velocity Obstacles). The simulation results show that the proposed scheme can deal with collision state in a short period time. Furthermore, the RNLVO computes the collisions using the trajectory of the robot. As a result, accurate prediction of the moving obstacles trajectory does not required.