• International Journal of Internet, Broadcasting and Communication
• /
• 제16권3호
• /
• pp.221-227
• /
• 2024

In autonomous navigation systems, the need for fast and accurate image processing using deep learning and advanced sensor technologies is paramount. These systems rely heavily on the ability to process and interpret visual data swiftly and precisely to ensure safe and efficient navigation. Despite the critical importance of such capabilities, there has been a noticeable lack of research specifically focused on ship image classification for maritime applications. This gap highlights the necessity for more in-depth studies in this domain. In this paper, we aim to address this gap by presenting a comprehensive comparative study of ship image classification using two distinct neural network models: the Feedforward Neural Network (FNN) and the Convolutional Neural Network (CNN). Our study involves the application of both models to the task of classifying ship images, utilizing a dataset specifically prepared for this purpose. Through our analysis, we found that the Convolutional Neural Network demonstrates significantly more effective performance in accurately classifying ship images compared to the Feedforward Neural Network. The findings from this research are significant as they can contribute to the advancement of core source technologies for maritime autonomous navigation systems. By leveraging the superior image classification capabilities of convolutional neural networks, we can enhance the accuracy and reliability of these systems. This improvement is crucial for the development of more efficient and safer autonomous maritime operations, ultimately contributing to the broader field of autonomous transportation technology.

• 제어로봇시스템학회논문지
• /
• 제16권9호
• /
• pp.898-904
• /
• 2010

Unmanned autonomous forklifts have a great potential to enhance the productivity of material handling in various applications because these forklifts can pick up and deliver loads without an operator and any fixed guide. There are, however, many technical difficulties in developing such forklifts including localization, map building, sensor fusion, control and so on. Implementation, which is often neglected, is one of practical issues in developing such an autonomous device. This is because the system requires numerous sensors, actuators, and controllers that need to be connected with each other, and the number of connections grows very rapidly as the number of devices grows. Another requirement on the integration is that the system should allow changes in the system design so that modification and addition of system components can be accommodated without too much effort. This paper presents a network-based distributed approach where system components are connected to a shared CAN network, and control functions are divided into small tasks that are distributed over a number of microcontrollers with a limited computing capacity. This approach is successfully applied to develop an unmanned forklift.

• 제어로봇시스템학회논문지
• /
• 제17권10호
• /
• pp.1051-1058
• /
• 2011

Unmanned autonomous forklifts have a great potential to enhance the productivity of material handling in various applications because these forklifts can pick up and deliver loads without an operator and any fixed guide. Especially, automation of pallet loading and unloading technique is useful for enhancing performance of logistics and reducing cost for automation system. There are, however, many technical difficulties in developing such forklifts including localization, map building, sensor fusion, control, and so on. This is because the system requires numerous sensors, actuators, and controllers that need to be connected with each other, and the number of connections grows very rapidly as the number of devices grows. This paper presents a vision sensorbased autonomous loading and unloading for network-based unmanned forklift where system components are connected to a shared CAN network. Functions such as image processing and control algorithm are divided into small tasks that are distributed over a number of microcontrollers with a limited computing capacity. And the experimental results show that proposed architecture can be an appropriate choice for autonomous loading in the unmanned forklift.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.45-45
• /
• 2000

This paper proposes a remote control system that combines computer network and an autonomous mobile robot. We control remotely an autonomous mobile robot with vision via the internet to guide it under unknown environments in the real time. The main feature of this system is that local operators need a World Wide Web browser and a computer connected to the internet communication network and so they can command the robot in a remote location through our Home Page. The hardware architecture of this system consists of an autonomous mobile robot, workstation, and local computers. The software architecture of this system includes the server part for communication between user and robot and the client part for the user interface and a robot control system. The server and client parts are developed using Java language which is suitable to internet application and supports multi-platform. Furthermore, this system offers an image compression method using motion JPEG concept which reduces large time delay that occurs in network during image transmission.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.67-72
• /
• 2023

In the past decade, Autonomous Vehicle Systems (AVS) have advanced at an exponential rate, particularly due to improvements in artificial intelligence, which have had a significant impact on social as well as road safety and the future of transportation systems. The fusion of light detection and ranging (LiDAR) and camera data in real-time is known to be a crucial process in many applications, such as in autonomous driving, industrial automation and robotics. Especially in the case of autonomous vehicles, the efficient fusion of data from these two types of sensors is important to enabling the depth of objects as well as the classification of objects at short and long distances. This paper presents classification of objects using CNN based vision and Light Detection and Ranging (LIDAR) fusion in autonomous vehicles in the environment. This method is based on convolutional neural network (CNN) and image up sampling theory. By creating a point cloud of LIDAR data up sampling and converting into pixel-level depth information, depth information is connected with Red Green Blue data and fed into a deep CNN. The proposed method can obtain informative feature representation for object classification in autonomous vehicle environment using the integrated vision and LIDAR data. This method is adopted to guarantee both object classification accuracy and minimal loss. Experimental results show the effectiveness and efficiency of presented approach for objects classification.

• 제어로봇시스템학회논문지
• /
• 제18권8호
• /
• pp.706-712
• /
• 2012

This paper presents a sensor network and operation for multiple autonomous navigation platform and transport service. Multiple platform navigate with inside sensors and outside sensors while acquiring and process some useful information. Each platform communicates each other by navigational information through central main server. Efficient sensor network systems are considered for the scenario which some passengers call the service and the vehicle accomplish its transport service by transporting each caller to the destination by autonomous manners. In the scenario, all vehicles perform a role of sensor system to the central server and the server handles each information and integrate with faster procedure in the wireless 3G network.

• 제어로봇시스템학회논문지
• /
• 제11권11호
• /
• pp.930-935
• /
• 2005

In this paper, position tracking control of an autonomous helicopter is presented. Combining an LQR method and a proportional control forms a simple PD control. Since LQR control gains are set for the velocity control of the helicopter, a position tracking error occurs. To minimize a position tracking error, neural network is introduced. Specially, in the frame of the reference compensation technique for teaming neural network compensator, a position tracking error of an autonomous helicopter can be compensated by neural network installed in the remotely located ground station. Considering time delay between an auto-helicopter and the ground station, simulation studies have been conducted. Simulation results show that the LQR with neural network performs better than that of LQR itself.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.94-99
• /
• 1993

In this paper, the method for navigation and obstacle avoidance of an autonomous mobile robot is proposed. It is based on the fuzzy inference system which enables to deal with imprecise and uncertain information, and on the neural network which enables to learn input and output pattern data obtained from ultrasonic sensors. For autonomous navigation, the wall-following navigation utilizing input and output data by an expert's control action is constructed. An approach by the neural network is developed for the obstacle avoidance because of the redundant input data. For an autonomous navigation, the fuzzy control and the control of the neural network are integrated and its feasibility is demonstrated by means of experiment.

• 대한임베디드공학회논문지
• /
• 제7권5호
• /
• pp.247-253
• /
• 2012

This paper presents sensor network based autonomous vehicle system using a proposed image processing algorithm. The proposed image processing algorithm consists of pre-processing and five-stage image processing: coordinate calculation, driving area decision, line segment calculation, steeling decision, and acceleration decision. We evaluate the performance of the proposed algorithm on both straight road and curved road. Experimental results indicate that the proposed algorithm works well for autonomous vehicles. However, control accuracy of the proposed algorithm decreases as speed is increasing.

• 융합보안논문지
• /
• 제19권1호
• /
• pp.19-30
• /
• 2019

최근 4차 산업혁명이 화두로 등장하면서 자율주행자동차의 중요성과 관심이 높아지고 있다. 전 세계적으로 시험 운행이 늘어나면서 자율주행자동차와 관련된 사고도 발생하고 있으며, 이에 대한 사이버 해킹 위협 가능성도 높아지고 있다. 미국, 영국, 독일을 포함한 여러 국가들은 이러한 추세를 반영하여 자율주행자동차의 사이버 해킹에 대응하기 위한 가이드라인을 만들거나 기존의 법률을 개정하고 있다. 국내의 경우 자율주행자동차의 제한적인 임시운행이 이루어지고 있으나, 자율주행자동차 해킹으로 인한 사고 발생 시 적용할 법제가 미흡한 상태이다. 본고에서는 기존의 관련 법률 체계를 분석하고 이를 바탕으로 자율주행자동차 사이버 해킹으로 인한 민사, 형사, 행정 책임 문제를 살펴보면서, 자율주행자동차 특성에 맞는 사고 책임 관련 법률체계를 제안하고 각 법제의 구성요소에 대해서 분석하여 이슈사항을 도출하며, 추가적으로 간략한 개선방안도 제시한다.