• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.6
• /
• pp.916-921
• /
• 2018

Recently, interest in autonomous vehicle has increased, and autonomous driving capability is also increasing. Depending on the autonomous driving ability, the maximum number of steps is divided into 6 steps. The higher the step, the less the involvement of the person in the running, and the person does not need to be involved at the highest stage. Today's autonomous vehicles have been developed high level, but solutions are not clearly defined in case of an accident, so only the test run is possible. Such an accident occurring during traveling is almost inevitable, and it is judged who has made a mistake in case of an accident, which increases the punishment for the accident. Although a black box is used to clarify such a part, it is easy to delete a record, and it is difficult to solve an accident such as a hit-and-run. In this paper, i design a driving record system using black chain to solve accidents.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.9
• /
• pp.369-374
• /
• 2014

Path generation methods are required for safe and efficient driving in unmanned autonomous vehicles. There are two kinds of paths: global and local. A global path consists of all the way points including the source and the destination. A local path is the trajectory that a vehicle needs to follow from a way point to the next in the global path. In this paper, we propose a novel method for local path generation through machine learning, with an effective curve function used for initializing the trajectory. First, reinforcement learning is applied to a set of candidate paths to produce the best trajectory with maximal reward. Then the optimal steering angle with respect to the trajectory is determined by training an artificial neural network. Our method outperformed existing approaches and successfully found quality paths in various experimental settings, including the cases with obstacles.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.6
• /
• pp.230-245
• /
• 2023

The importance of evaluating the safety of Automated Vehicles (AV) is increasing with the advances in autonomous driving technology. Accordingly, an evaluation scenario that defines in advance the situations AV may face while driving is being used to conduct efficient stability evaluation. On the other hand, the single scenarios currently used in conventional evaluations address limited situations within short segments. As a result, there are limitations in evaluating continuous situations that occur on real roads. Therefore, this study developed a set of multiple scenarios that allow for continuous evaluation across entire sections of roads with diverse geometric structures to assess the safety of AV. In particular, the conditions for connecting individual scenarios were defined, and a methodology was proposed for developing concrete multiple scenarios based on the scenario evaluation procedure of the PEGASUS project. Furthermore, a simulation was performed to validate the practicality of these multiple scenarios.

• TTA Journal
• /
• s.173
• /
• pp.35-40
• /
• 2017

• The Proceeding of the Korean Institute of Electromagnetic Engineering and Science
• /
• v.28 no.4
• /
• pp.27-36
• /
• 2017

• Journal of Drive and Control
• /
• v.19 no.1
• /
• pp.77-80
• /
• 2022

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.561-568
• /
• 2021

In recent years, autonomous driving technologies have become a high-value-added technology that attracts attention in the fields of science and industry. For smooth Self-driving, it is necessary to accurately detect an object and estimate its movement speed in real time. CNN-based deep learning algorithms and conventional dense optical flows have a large consumption time, making it difficult to detect objects and estimate its movement speed in real time. In this paper, using a single camera image, fast object detection was performed using the YOLOv5 algorithm, a deep learning algorithm, and fast estimation of the speed of the object was performed by using a local dense optical flow modified from the existing dense optical flow based on the detected object. Based on this algorithm, we present a system that can predict the collision time and probability, and through this system, we intend to contribute to prevent tram accidents.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.4
• /
• pp.391-399
• /
• 2020

Autonomous driving can be limited by only using sensors if the sensor is blocked by sudden changes in surrounding environments or large features such as heavy vehicles. In order to overcome the limitations, the precise road-map has been used additionally. In korea, the NGII (National Geographic Information Institute) produces and supplies high definition map for autonomous vehicles. Accordingly, in this study, errors occurring in the process of e data editing and dtructured esditing of high definition map are systematically typed providing by the National Geographic Information Institute. In addition, by presenting the error search process and solution for each situation, we conducted a study to quickly correct errors in high definition map, and largely classify the error items for shape integrity, spatial relationship, and reference relationship, and examine them in detail. The method was derived.

• Journal of Internet Computing and Services
• /
• v.20 no.5
• /
• pp.105-111
• /
• 2019

Core technologies that lead the Fourth Industrial Revolution era, such as artificial intelligence, big data, and autonomous driving, are implemented and serviced through the rapid development of computing power and hyper-connected networks based on the Internet of Things. In this paper, we implement two different models for drivable area segmentation in various environment, and propose a better model by comparing the results. The models for drivable area segmentation are using DeepLab V3+ and Mask R-CNN, which have great performances in the field of image segmentation and are used in many studies in autonomous driving technology. For driving information in various environment, we use BDD dataset which provides driving videos and images in various weather conditions and day&night time. The result of two different models shows that Mask R-CNN has higher performance with 68.33% IoU than DeepLab V3+ with 48.97% IoU. In addition, the result of visual inspection of drivable area segmentation on driving image, the accuracy of Mask R-CNN is 83% and DeepLab V3+ is 69%. It indicates Mask R-CNN is more efficient than DeepLab V3+ in drivable area segmentation.

• Journal of the KSME
• /
• v.34 no.10
• /
• pp.764-769
• /
• 1994

본 연구에서 개발한 KARV-2의특징은 저가의 소형시스템이다. 4인승 소형지프차에 모든 컨트 롤시스템을 장착하고도 외부에 별로 표가 나지 않을 정도임에도 그 성능 면에서는 고속도로를 자율주행할 수 있다는 점이다. 첨단기술의 개발과 더불어 이 차량의 성능도 계속 개선될 것이다. 언젠가는 궤도 위를 달리는 기차와 같이 정확하고 안전하게 주행할 수 있어, 탑승자는 마음놓고 그저 신문을 읽거나 음료수를 마시면서 스쳐 가는 경관을 즐길 수 있는 시기가 곧 도래할 것 으로 믿는 다. 이를 위하여는 차량작동의 신뢰도 향상이 절대 전제조건이다. 앞으로는 일기조건 이나 주야에 관계없이 마치궤도 위를 달리는 기차처럼 안전하게 주행할 수 있는 시스템의 개발이 필요하다. 또한 고속도로가 아닌 시내주행을 위해서는 노면표식인식, 도로표지판 인식, 신호등 인식 등 해결되어야 할 문제들이 산적해 있는 상황이다.