• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.246-248
• /
• 2017

Self-driving cars have three main functions. The first recognizes the surrounding environment, judge the risk, and lastly plans the drive path. Therefore, the driving operation is minimized. And it refers to a human friendly car capable of safe driving on its own. The reason for the need for self-driving car was to reduce traffic jams on limited roads and to reduce carbon dioxide emissions. Driving ahead of these self-driving car businesses can be expected to attract and expand the existing business and expand the new business and create new business opportunities for ICT firms. It is urgent for the concerned agencies to establish legal and institutional basis for self-driving cars. By doing so, new services could be provided to consumers. Therefore, this paper introduces the technological development trends for self-driving cars.

• Journal of the Korea Convergence Society
• /
• v.9 no.5
• /
• pp.7-14
• /
• 2018

The purpose of this study is to analyze the effect of commuter's departure time on transportation system in future traffic system operated autonomous vehicle using agent based model. Various scenarios have been set up, such as when all passenger choose a similar departure time, or if the passenger chooses a different departure time. Also, this study tried to analyze the effect of road capacity. It was found that although many of the scenarios had been completed in a stable manner, many commuters were significantly coordinated at the desired departure time. In particular, in the case of a reduction in road capacity or in certain scenarios, it has been shown that, despite excessive schedule adjustments, many passengers are unable to commute before 9 o'clock. As a result, it is suggested that traffic management and pricing policies are different from current ones in the era of autonomous car operation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.10
• /
• pp.1294-1299
• /
• 2018

'SMART Police' means collecting, analyzing and utilizing the police database such as crime statistics, reflecting 'strategic management', 'analysis and research', 'science technology' do. In recent years, the number of sex offenses has more than doubled based on the 'crime analysis' of the Supreme Prosecutors' In this trend, we started to build a sex crime prevention program for women 's relief special 3.0 which is suitable for regional characteristics. However, it was not enough for the women to have low publicity and utilization results, and to show the practical effect. In the future, we propose a crime prevention system which can safely take home safeguarding with autonomous driving car or drone by applying ICT to the future police system. Although it is difficult to operate autonomous vehicles and drones in the surrounding environment such as narrow alleys and power lines for application to this system, it is thought that the artificial intelligence technology can be sufficiently overcome to operate.

• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we propose an autonomous driving system using an end-to-end model to improve lane departure and misrecognition of traffic lights in a vision sensor-based system. End-to-end learning can be extended to a variety of environmental conditions. Driving data is collected using a model car based on a vision sensor. Using the collected data, it is composed of existing data and data with outlayers removed. A class was formed with camera image data as input data and speed and steering data as output data, and data learning was performed using an end-to-end model. The reliability of the trained model was verified. Apply the learned end-to-end model to the model car to predict the steering angle with image data. As a result of the learning of the model car, it can be seen that the model with the outlayer removed is improved than the existing model.

• Smart Media Journal
• /
• v.11 no.9
• /
• pp.9-20
• /
• 2022

Deep reinforcement learning(RL) is an end-to-end data-driven control method that is widely used in the autonomous driving domain. However, conventional RL approaches have difficulties in applying it to autonomous driving tasks due to problems such as inefficiency, instability, and uncertainty. These issues play an important role in the autonomous driving domain. Although recent studies have attempted to solve these problems, they are computationally expensive and rely on special assumptions. In this paper, we propose a new algorithm MCDT that considers inefficiency, instability, and uncertainty by introducing a method called uncertainty sequence modeling to autonomous driving domain. The sequence modeling method, which views reinforcement learning as a decision making generation problem to obtain high rewards, avoids the disadvantages of exiting studies and guarantees efficiency, stability and also considers safety by integrating uncertainty estimation techniques. The proposed method was tested in the OpenAI Gym CarRacing environment, and the experimental results show that the MCDT algorithm provides efficient, stable and safe performance compared to the existing reinforcement learning method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.10
• /
• pp.274-280
• /
• 2017

The smart-car industry has emerged as the important variable that will decide the future industrial contour of the automotive industry, together with commercialization of electronic vehicles, connected cars, infotainment, telematics, and the autonomous/self-driving car. This study analyzes the strategic position of platform companies and car manufacturers that would determine the future of the smart-car market. The findings of this study show that despite the entry barriers in industrial factors, such as economies of scale, the industrial infrastructure, and global production networks, and technical factors like exclusive head-sector information, car manufacturers may be deprived of their industrial leadership by platform companies with map and user data, big data capabilities, and user interface experience if they lag behind ICT innovation. This insight is based on the emerging importance of software and platforms, and the simplification of car structures, proven by the successful commercialization of electronic vehicles. This study complements existing studies mainly focused on technical aspects of the smart-car industry by examining the strategic dimensions of platform companies and their approach to the future smart-car market by comparing them with existing car manufacturing multinationals.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.8 no.5
• /
• pp.249-254
• /
• 2013

In this paper we propose nonlinear control system for automatic unmanned vehicle using a RC (Radio Controlled) car which is usually controlled by a remote controller. In the proposed system, a RC car is dissembled and reassembled with several parts enabling it to be controlled by an android mobile platform with Bluetooth communication. In our system, an android mobile smartphone is mounted on the RC car and plays an important role as an eye of the car. The proposed system automatically controls the RC car to follow a lane that we draw on the floor of our laboratory. Also, the proposed RC car system can also be controlled manually using the accelerometer sensor of a smartphone through a Bluetooth module. Our proposed system that has both manual mode and automatic mode consists of several components; a microprocessor unit, a Bluetooth serial interface module, a smartphone, a dual motor controller and a RC toy car. We are now in the development of a group driving system in which one car follows the front car that tracks a lane automatically.

• Electronics and Telecommunications Trends
• /
• v.38 no.6
• /
• pp.31-40
• /
• 2023

Given recent accidents involving autonomous vehicles, driver monitoring technology related to the transition of control in autonomous vehicles is gaining prominence. Driver status monitoring systems recognize the driver's level of alertness and identify possible impairments in the driving ability owing to conditions including drowsiness and distraction. In autonomous vehicles, predictive factors for the transition to manual driving should also be included. During traditional human driving, monitoring the driver's status is relatively straightforward owing to the consistency of crucial cues, such as the driver's location, head orientation, gaze direction, and hand placement. However, monitoring becomes more challenging during autonomous driving because of the absence of direct manual control and the driver's engagement in other activities, which may obscure the accurate assessment of the driver's readiness to intervene. Hence, safety-ensuring technology must be balanced with user experience in autonomous driving. We explore relevant global and domestic regulations, the new car assessment program, and related standards to extract requirements for driver status monitoring. This kind of monitoring can both enhance the autonomous driving performance and contribute to the overall safety of autonomous vehicles on the road.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.349-350
• /
• 2019

The autonomous driving drones experimented in the existing indoor corridor environment was a way to give the steering command to the drones by the neural network operation of the notebook due to the limitation of the operation performance of the drones. In this paper, to overcome these limitations, we have studied autonomous driving in indoor corridor environment using NVIDIA Jetson TX2 board.

• Journal of Auto-vehicle Safety Association
• /
• v.14 no.4
• /
• pp.60-69
• /
• 2022

Recently, the importance of simulation validation in a virtual environment for autonomous driving system validation is increasing. At the same time, interest in the advancement of the virtual driving environment is also increasing. To develop autonomous driving technology, a simulation environment similar to the real-world environment is needed. For this reason, not only the road model is configured in the virtual driving environment, but also the driving environment configuration that includes the surrounding environments -traffic, object, etc- is necessary. In this article, FMTC, which is a test bed for autonomous vehicles, is implemented in a virtual environment and advanced to form a virtual driving environment similar to that of real FMTC. In addition, the similarity of the virtual driving environment is verified through comparative analysis with the real FMTC.