• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.211-214
• /
• 2000

This paper describes a lateral guidance system of an autonomous vehicle, using a neural network model of magneto-resistive sensor and magnetic fields. The model equation was compared with experimental sensing data. We found that the experimental result has a negligible difference from the modeling equation result. We verified that the modeling equation can be used in simulations. As the neural network controller acquires magnetic field values(B$\sub$x/, B$\sub$y/, B$\sub$z/) from the three-axis, the controller outputs a steering angle. The controller uses the back-propagation algorithms of neural network. The learning pattern acquisition was obtained using computer simulation, which is more exact than human driving. The simulation program was developed in order to verify the acquisition of the teaming pattern, learning itself, and the adequacy of the design controller. Also, the performance of the controller can be verified through simulation.

• Journal of Sensor Science and Technology
• /
• v.12 no.5
• /
• pp.211-217
• /
• 2003

This paper is proposed that roadway recognition performance improvement for autonomous vehicle using magnetic markers that are embedded along the road center and the sensors mounted on a vehicle, and which changing of magnetic field that is measured along with vehicle driving. For Retrenchment of equipment cost, interval of markers is more expensive than existing method. In order to this, This paper is proposed that interval of markers is founded using magnetic field analysis, and which arrangement method of six magnetic sensors and control method of neural network. This paper is carried out magnetic field analysis, the acquiring of the training patterns, the training of the neural network and composition of steering control, and is verified that roadway recognition performance can improve using computer simulation with proposed methods.

• Journal of Korea Society of Industrial Information Systems
• /
• v.26 no.4
• /
• pp.11-18
• /
• 2021

We propose a method that combines learning in a virtual environment and a real environment for indoor autonomous driving through reinforcement learning. In case of learning only in the real environment, it takes about 80 hours, but in case of learning in both the real and virtual environments, it takes 40 hours. There is an advantage in that it is possible to obtain optimized parameters through various experiments through fast learning while learning in a virtual environment and a real environment in parallel. After configuring a virtual environment using indoor hallway images, prior learning was carried out on the desktop, and learning in the real environment was conducted by connecting various sensors based on Jetson Xavier. In addition, in order to solve the accuracy problem according to the repeated texture of the indoor corridor environment, it was possible to determine the corridor wall object and increase the accuracy by learning the feature point detection that emphasizes the lower line of the corridor wall. As the learning progresses, the experimental vehicle drives based on the center of the corridor in an indoor corridor environment and moves through an average of 70 steering commands.

• The Journal of Fisheries Business Administration
• /
• v.48 no.3
• /
• pp.33-45
• /
• 2017

There have been several complications and problems about marine ranching policy. Regional conflicts have occurred in the process of selecting land for the pasture. And conflict occurred in post-management process. The causes of these conflicts were pointed out to lack of social science research and alternatives. The government-led initiative model, which ignores participation in each sector, has been pointed out as a problem. In order to increase the effectiveness of the marine ranching project, the following points should be considered. First, at the stage of establishing the ocean ranch business plan, social science approach should be considered together with natural science approach. Second, a governance system should be established to enhance the effectiveness of the marine ranch project. The governance system refers to a consultative body in which the central government and local governments, autonomous steering committees, research institutes, local civil society organizations, and participating companies participate. We can improve the efficiency of business by establishing and promoting direction, operation policy, and action plan of sea pasture business centering on governance system. Third, it is necessary to change the management system from the existing administration-led business model to the cooperative governance system - based model. The effectiveness of the marine ranching project can be improved through the governance system.

• Journal of Navigation and Port Research
• /
• v.44 no.2
• /
• pp.136-143
• /
• 2020

Maritime autonomous surface ships (MASS) has been developed over the years. But, there are many unresolved problems. To overcome these problems, this study proposes connected ship navigation system. The system comprises a slave ship and a master ship that leads the slave ship. To implement this system, communication network, route planning algorithms, and controllers are designed. The communication network is built using the transmission control protocol/Internet protocol (TCP/IP) socket communication method to exchange data between ships. The route planning algorithms calculate the course and distance of the slave ship using the middle latitude sailing method. Nomoto model is used as the mathematical model of the slave ship maneuvering motion. Then, the autoregressive with exogenous variables (ARX) model is used to estimate the parameters of Nomoto model. Based on the above model, the automatic steering controller is designed using a proportional-derivative (PD) control. Also, the speed controller is designed for the slave ship to maintain constant distance from the master ship. Sea experiments are conducted to verify the proposed system with two remodeled boats.

• Journal of Ocean Engineering and Technology
• /
• v.20 no.6 s.73
• /
• pp.24-34
• /
• 2006

Hydrodynamic coefficients strongly affect the dynamic performance of an AUV. Thus, it is important to know the true values of these coefficients, in order to accurately simulate the AUV's dynamic performance. Although these coefficients are generally obtained experimentally, such as through the PMM test, the measured values are not completely reliable because of experimental difficulties and errors. Another approach, by which these coefficients can be obtained, is the observer method, in which a model-based estimation algorithm estimates the coefficients. In this paper, the hydrodynamic coefficients are estimated using two nonlinear observers: a sliding mode observer and an extended Kalman filter. Their performances are evaluated in Matlab simulations, by comparing the estimated coefficients obtained from the two observer methods, with the experimental values as determined from the PMM test. A sliding mode controller is constructed for the diving and steering maneuver by using the estimated coefficients. It is demonstrated that the controller, applied with the estimated values, maintains the desired depth and path with sufficient accuracy.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.2
• /
• pp.185-190
• /
• 2016

This paper deals with analysis of the characteristics of mobile robot's motion by automatic detection of markers on a robot using a camera. Analysis of motion behaviors according to parameters is important in developing control algorithm for robot operation or autonomous navigation. For this purpose, we use four chessboard patterns on the robot. Their location on the robot is adjusted to be on single plane. Homography is used to compute the actual amount of movement of the robot. Presented method is tested using P3-AT robot and it gives reliable results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.6
• /
• pp.775-780
• /
• 2008

In this paper, unmanned driving of robotic vehicle using magnetic marker is proposed. One of the most important component of autonomous vehicle is to detect the position of a magnetic marker on the road. In order to calculate the precise position of a magnet embedded on the road, the relation of magnetic field and a sensor is analyzed, and a new position sensing system using arrayed magnetic sensor is proposed. Also, the steering control system using a stepping motor is developed for driving by automatic mode as well as manual mode. For the verification of usability, the developed robotic vehicle is tested on magnetic road.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.152-152
• /
• 2017

본 연구에서는 비례제어밸브를 이용한 자율주행 트랙터 조향 시스템 제어를 위하여 저가의 임베디드 시스템을 사용한 제어기를 개발하였다. 차륜의 조향각 측정을 위하여 전륜 킹핀에 포텐시오미터를 설치하였으며, 비례제어밸브는 -10 ~ +10V의 전압으로 밸브 스풀의 위치제어를 수행하도록 하였다. 조향각 측정과 비례제어밸브의 위치제어를 위하여 각각에 AT90CAN128 AVR보드를 사용하였으며, CAN통신으로 조향각 데이터가 비례제어밸브 제어용 데이터로 전송될 수 있도록 하였다. 비례제어밸브 제어 보드에는 DAC기능을 추가하였으며 0 ~ 5V의 출력을 -10 ~ +10V의 전압으로 변환해 주는 인터페이스회로를 추가하였다. 일반적으로 GPS 등의 데이터 수신율이 20 Hz인 점을 감안하여 비례제어 밸브는 50 ms의 주기로 P-제어를 수행할 수 있도록 하였다. 향후 트랙터의 방향각을 설정하는 또 하나의 시스템으로부터 목표 조향각을 부여받는 것을 가상하여 별도의 MCU를 통해 목표 조향각을 송신한 후, 조향 유압실린더에 의한 전륜의 조향각 시간 응답 특성을 조사하였다. 실험은 트랙터의 전륜을 지면으로부터 들어올린 무 부하 상태에서 진행하였으며, 목표 조향각은 $7.5^{\circ}$, $15.0^{\circ}$, $22.5^{\circ}$ 등 3단계로 변경하며 시간응답 특성을 측정하였다. 최대 오버슈트 11%, 최소 오버슈트 8.6%, 정상상태 오차 약 $0.825^{\circ}$, 시정수(Time Constant)는 3단계의 목표 조향각 설정에서 각각 0.706초, 0.488초, 0.38초로 나타났다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.6
• /
• pp.584-593
• /
• 2012

This paper introduces an AVP (Automated Valet Parking) system which applies an autonomous driving concept into the current PAS (Parking Assistant System). The present commercial PAS technology is limited into vehicle. It means vehicle only senses and controls by and for itself to assist the parking. Therefore, the present PAS is restricted to simple parking events. But AVP includes wider parking events and planning because it uses infra-sensor network as well as vehicle sensor. For the realization of AVP, the commercial steering system of a compact vehicle was modified into steer-by-wire structure and various sensors like LRF (Long Range Finder) and camera were installed in a parking area. And local & global server decides where and when the vehicle can go and park in the testing area after recognized the status of environment and vehicle from those sensors. GPS solution was used to validate the AVP performance. More various parking situations, vehicles and obstacles will be considered in the next research stages based on these results. And we expect this AVP solution with more intelligent vehicles can be applied in a big parking lot like a market, an amusement park, etc.