http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
李基聖,신영국 弘益大學校 科學技術硏究所 2000 科學技術硏究論文集 Vol.11 No.-
In the path planning for a mobile robot, Distance Transform(DT), Certainty Grid Map and Artificial Intelligent algorithms are widely used. The above three algorithm have several drawbacks to generate the best path of all possible paths. In this paper new path planning algorithms are proposed. Three methods of path planning are Primitive Transform, Fast Primitive Distance Transform, Land Gambling Distance Transform. The simulation is performed to show that the proposed algorithms can generate the best path for a mobile robot.
李基聖,郭漢澤 弘益大學校 科學技術硏究所 1995 科學技術硏究論文集 Vol.6 No.-
A 3-D object image recognition and restoration using the ultrasound array sensor is presented. To obtain an image of the object, the planar arrangement of the ultrasound sensor is used. The acquired data are learned by the SCL(Simple Competitive Learning) Neural Network. Lab experiments were performed to show that the object can be recognized by using small number of the ultrasound sensor and moment method.
거리변환법과 벡터장 히스토그램을 이용한 자율주행 로봇의 경로계획
李基聖 弘益大學校 科學技術硏究所 1994 科學技術硏究論文集 Vol.4 No.-
A real-time path plan for an AMR (Autonomous Mobile Robot) using the DT(Distance Transform) and VFH(Vector Field Histogram) is studied. This approach enables AMR to find a path with the known environment and unknown obstacles simultaneously by using sensors for avoiding collisions and moving toward the goal. The developed software provides a GUI(Graphics User Interface) to use the AMR system easily. All system parameters including positions of the obstacles can be accepted using a mouse icon and all the obtained trajectories can be displayed on a computer monitor in graphics or file forms.
李基聖,김용길 弘益大學校 科學技術硏究所 1998 科學技術硏究論文集 Vol.9 No.2
During the navigation of a mobile robot, one of the essential task is to plan the optimal path from a starting point to a destination point. This paper describes the global path planner using HNN(Hopfield Neural Network) which can be realized in real time and with a high reliability. Using Hopfield type neural network, it is needed to choose the energy coefficient. Usually, it is determined by trial and errors. In this paper, in the beginning of the simulation the weight factor is selected as a large value, and the factor is diminished. The simulation results show accuracy and efficiency of the proposed algorithm.
李基聖 弘益大學校 科學基術硏究所 1999 科學技術硏究論文集 Vol.10 No.2
In order to make an autonomous mobile robot, techniques of automatic position estimation, artificial intelligence control using sensor needed. Especially, automatic position estimation technique is essential. The autonomous mobile robot estimates the position using encoder readings. Generally, the conventional PI-controller with fixed parameters is used to compensate the error of mobile robot. To reduce the error between the desired trajectory and the measured trajectory, the parameters in PI is tamed using a genetic algorithm. A new method using a genetic algorithm to find optimal PI coefficients is proposed. The position and angle errors that resulted from unbalanced wheel-loads of a mobile robot with encoder from wheels are measured. And in oder to find optimal PI coefficients, the genetic algorithm whose fitness function is established to minimize errors is used.
자율주행 운반체의 기구학적 모델과 자기 동조 제어기 알고리즘 연구
李基聖 弘益大學校 科學技術硏究所 1993 科學技術硏究論文集 Vol.3 No.-
The kinematic model and self-tuning algorithm of a mobile robot is studied. The driving and steering motor assembly is located in the front of the mobile robot. The position of the mobile robot is determined by the steering angle and driving distance. For the controller design, a time-series multivariate model of the autoregressive exogenous (ARX) type is used to describe the input-output relation. The discounted least square method is used to estimate parameters of the time-series model. A self-tuning controller is so designed that the position of the center of the mobile robot track the given trajectory. Simulation result controlled by a self-tuning controller is presents to illustrate the approach.
李基聖 弘益大學校 科學技術硏究所 2005 科學技術硏究論文集 Vol.16 No.-
For the factory automations and flexible manufacturing systems, mobile robot systems are introduced. To be used in the real factory environments, the capability of the position sensing, collision avoidance and control technology using the sensors and navigation is needed. Navigation is a method to direct a mobile robot without collision when traversing known or unknown environments. In this paper, new algorithm for the path planning and collision avoidance is proposed. The genetic algorithm searches for a path in the entire and continuous free space and unifies global path planning and local path planning. The simulation result shows the proposed method is an efficient and effective method when compared with the traditional collision avoidance algorithms.