• 자동차안전학회지
• /
• 제11권4호
• /
• pp.39-49
• /
• 2019

Autonomous vehicle is a car which drives itself without any human interaction. SAE provides technical definitions for autonomous and international standards for test evaluation. Accordingly, automobile industry is actively researching development and evaluation of various ADAS (Advanced Driver Assistance Systems), : representative technology of autonomous technology. Recently, ADAS is in the commercialization level such as ACC, LKAS, AEB, and HDA etc. And it also has issues about safety evaluation. The purpose of HDA in ADAS is reduced the driving load on highway. It has a function which can maintain lane keeping and control distance from forward vehicle. This function is evaluated to be useful for accident prevention. Therefore, this paper proposes the safety evaluation scenario of HDA, considering the domestic highway design criteria and the situation that may arise on the actual highway. We compared and analyzed the data acquired through simulation and actual vehicle test. And verified the reliability of the proposed safety evaluation scenario. The verified result is expected safety evaluation of HDA is possible even under the bad condition, which cannot be tested.

• 로봇학회논문지
• /
• 제19권1호
• /
• pp.130-138
• /
• 2024

High-precision 3D Object Detection is a crucial component within autonomous driving systems, with far-reaching implications for subsequent tasks like multi-object tracking and path planning. In this paper, we propose a novel approach designed to enhance the performance of 3D Object Detection, especially in heading angle estimation by employing a moving object segmentation technique. Our method starts with extracting point-wise moving labels via a process of moving object segmentation. Subsequently, these labels are integrated into the LiDAR Pointcloud data and integrated data is used as inputs for 3D Object Detection. We conducted an extensive evaluation of our approach using the KITTI-road dataset and achieved notably superior performance, particularly in terms of AOS, a pivotal metric for assessing the precision of 3D Object Detection. Our findings not only underscore the positive impact of our proposed method on the advancement of detection performance in lidar-based 3D Object Detection methods, but also suggest substantial potential in augmenting the overall perception task capabilities of autonomous driving systems.

• 자동차안전학회지
• /
• 제16권1호
• /
• pp.6-11
• /
• 2024

This paper presents a motion planning algorithm of autonomous racing vehicles for mimicking the characteristics of a human driver. Time optimal maneuver of a race car has been actively studied as a major research area over the past decades. Although the time optimization problem yields a single time series solution of minimum time maneuver inputs for the vehicle, human drivers achieve similar lap times while taking various racing lines and velocity profiles. In order to model the characteristics of a specific driver and reproduce the motion, a stochastic motion planning framework based on kernelized motion primitive is introduced. The proposed framework imitates the behavior of the generated reference motion, which is based on a small number of human demonstration laps along the racetrack using Gaussian mixture model and Gaussian mixture regression. The mean and covariance of the racing line and velocity profile mimicking the driver are obtained by accumulating the outputs tested at equidistantly sampled input points. The results confirmed that the obtained lateral and longitudinal motion simulates the driver's driving characteristics, which are feasible for actual vehicle test environments.

• 대한토목학회논문집
• /
• 제44권1호
• /
• pp.63-73
• /
• 2024

현재 전 세계적으로 연구·개발 중인 자율주행차 Level 3에서 Level 5단계는 운전자의 인지-판단-제어과정을 차량에 탑재된 각종 센서로 대체하여, 운전과정의 대부분을 인공지능이 자율적으로 수행할 수 있도록 한다. 하지만 현재 자율주행차는 국가별로 상이한 자율주행차의 판단능력 최소기준을 만족할 경우, 임시운행 허가를 받아 도로주행이 가능하도록 하고 있다. 향후 자율주행차가 보급될 때 구매자들은 임시운행 허가의 한계로 위험상황 회피능력에 대한 신뢰도가 높지 않을 것으로 예상된다. 이에 본 연구에서는 드라이빙 시뮬레이터 기반으로 운전자의 위험상황 회피능력 비교·평가를 통해 자율주행차 판단능력 등급화 방안 제시 및 시나리오별 등급화가 가능한 평가지표를 도출하고자 하였다. 드라이빙 시뮬레이터 실험에는 성인 30명(남=25, 여=5명)이 참여하였다. 실험결과 분석은 K-평균 군집분석과 독립표본 T-검정을 진행하였으며, 이를 통해 자율주행차의 판단능력 등급 분류가 가능함과 등급 분류의 통계적 유의성을 확인할 수 있었다. 향후 자율주행차의 위험상황 회피능력에 대한 신뢰수준을 향상시키는데 크게 기여할 수 있을 것이다.

• 로봇학회논문지
• /
• 제12권3호
• /
• pp.297-305
• /
• 2017

As the development of autonomous vehicles becomes realistic, many automobile manufacturers and components producers aim to develop 'completely autonomous driving'. ADAS (Advanced Driver Assistance Systems) which has been applied in automobile recently, supports the driver in controlling lane maintenance, speed and direction in a single lane based on limited road environment. Although technologies of obstacles avoidance on the obstacle environment have been developed, they concentrates on simple obstacle avoidances, not considering the control of the actual vehicle in the real situation which makes drivers feel unsafe from the sudden change of the wheel and the speed of the vehicle. In order to develop the 'completely autonomous driving' automobile which perceives the surrounding environment by itself and operates, ability of the vehicle should be enhanced in a way human driver does. In this sense, this paper intends to establish a strategy with which autonomous vehicles behave human-friendly based on vehicle dynamics through the reinforcement learning that is based on Q-learning, a type of machine learning. The obstacle avoidance reinforcement learning proceeded in 5 simulations. The reward rule has been set in the experiment so that the car can learn by itself with recurring events, allowing the experiment to have the similar environment to the one when humans drive. Driving Simulator has been used to verify results of the reinforcement learning. The ultimate goal of this study is to enable autonomous vehicles avoid obstacles in a human-friendly way when obstacles appear in their sight, using controlling methods that have previously been learned in various conditions through the reinforcement learning.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 춘계학술대회
• /
• pp.683-686
• /
• 2018

자율주행자동차는 인공지능과 초 연결기술을 활용한 4차 산업혁명에 해당하며, 전 세계적으로 많은 투자와 연구가 진행되고 있는 사업이다. 그러나 지난 3월 미국 애리조나에서 시험운행 중인 우버 차량이 어두운 밤에 길을 건너던 보행자 들이받아 사망한 사고가 발생하고 지난 4월 테슬라 차량이 태양의 역광이 내리쬐는 상황에서 잘못된 판단으로 중앙 분리대를 들이받는 사고가 연속적으로 발생하였다. 이러한 문제들은 자율주행자동차에 탑재된 센서가 눈 비 태양광 등 악천후에 따른 잘못된 인식과 판단으로 발생한 사고들이였다. 본 논문에서는 자율주행자동차의 구성과 사건의 원인을 분석하고 인명사고가 발생할 수 있는 위급 상황에서 판단해야할 기준에 대하여 생각해 보았다.

• 로봇학회논문지
• /
• 제17권2호
• /
• pp.198-208
• /
• 2022

In this paper, an autonomous driving system is implemented for the Segye AI Robot Race Competition that multiple vehicles drive simultaneously. By utilizing the ERP42-racing platform, RTK-GPS, and LiDAR sensors provided in the competition, we propose an autonomous driving system that can drive safely and quickly in a road environment with multiple vehicles. This system consists of a recognition, judgement, and control parts. In the recognition stage, vehicle localization and obstacle detection through waypoint-based LiDAR ROI were performed. In the judgement stage, target velocity setting and obstacle avoidance judgement are determined in consideration of the straight/curved section and the distance between the vehicle and the neighboring vehicle. In the control stage, adaptive cruise longitudinal velocity control based on safe distance and lateral velocity control based on pure-pursuit are performed. To overcome the limited experimental environment, simulation and partial actual experiments were conducted together to develop and verify the proposed algorithms. After that, we participated in the Segye AI Robot Race Competition and performed autonomous driving racing with verified algorithms.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2004년도 추계학술대회 논문집
• /
• pp.908-911
• /
• 2004

In this paper, we describe a robust image processing algorithm to recognize the road lane in real-time. For the real-time processing, a detection area is decided by a lane segment of a previous frame and edges are detected on the basis of the lane width. For the robust driving, the global threshold with the Otsu algorithm is used to get a binary image in a frame. Therefore, reliable edges are obtained from the algorithms suggested in this paper in a short time. Lastly, the lane segment is found by hough transform. We made a RC(Radio Control) car equipped with a vision system and verified these algorithms using the RC Car.

• 대한안전경영과학회지
• /
• 제10권1호
• /
• pp.205-213
• /
• 2008

Servicizing is defined 'not as to sell the products but as to sell the functions'. This study was done on the methods for revitalization as the object items of navigation & antenna among car parts. The direct door-to-door survey was done for 101 enterprises (direct-management stores) and percentage & t-test were done for statistics. The results can be summarized like followings. First, PR on servicizing is required. Second the optimized methods can be obtained by incorporating reuse & recycling distribution systems of servicizing with the existing recovery system Reuse distribution system can be constructed rationally because inventory control is being done correct1y with the periodic recovery. Finally, as for the analysis on the relative importance of main factors of servicizing, it shows that supports & providing the concrete frame of the government & local autonomous body are important.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권12호
• /
• pp.5669-5684
• /
• 2018

The new mobile edge network architecture has been required for an increasing amount of traffic, quality requirements, advanced driver assistance system for autonomous driving and new cloud computing demands on highway. This article proposes a hierarchical cloud computing architecture to enhance performance by using adaptive data load distribution for buses that play the role of edge computing server. A vehicular dynamic cloud is based on wireless architecture including Wireless Local Area Network and Long Term Evolution Advanced communication is used for data transmission between moving buses and cars. The main advantages of the proposed architecture include both a reduction of data loading for top layer cloud server and effective data distribution on traffic jam highway where moving vehicles require video on demand (VOD) services from server. Through the description of real environment based on NS-2 network simulation, we conducted experiments to validate the proposed new architecture. Moreover, we show the feasibility and effectiveness for the connected car media service on highway.