• Journal of Drive and Control
• /
• v.19 no.4
• /
• pp.110-116
• /
• 2022

Recently, various studies have been conducted to apply deep learning and AI to various fields of autonomous driving, such as recognition, sensor processing, decision-making, and control. This paper proposes a controller applicable to path following, static obstacle avoidance, and pedestrian avoidance situations by utilizing reinforcement learning in autonomous vehicles. For repetitive driving simulation, a reinforcement learning environment was constructed using virtual environments. After learning path following scenarios, we compared control performance with Pure-Pursuit controllers and Stanley controllers, which are widely used due to their good performance and simplicity. Based on the test case of the KNCAP test and assessment protocol, autonomous emergency steering scenarios and autonomous emergency braking scenarios were created and used for learning. Experimental results from zero collisions demonstrated that the reinforcement learning controller was successful in the stationary obstacle avoidance scenario and pedestrian collision scenario under a given condition.

• Journal of IKEEE
• /
• v.22 no.1
• /
• pp.223-226
• /
• 2018

Recently, various safety technologies have been extensively developed to protect occupants from accidents. This paper surveys various automotive occupant protection technologies such as antilock braking system, traction control system, electronic brake distribution, electronic stability control, autonomous emergency braking, airbag, seatbelt pretensioner, and active headrest. Their operation principles and implementations are also explained.

• Journal of Auto-vehicle Safety Association
• /
• v.13 no.4
• /
• pp.106-113
• /
• 2021

In this study, The behavior of an autonomous vehicle in an intersection accident situation is predicted. Based on a representative intersection accident situation from actual intersection accident database, simulation was performed by applying the automatic emergency braking algorithm used in the autonomous driving system. Accident reconstruction was performed based on the accident report of the representative accident situation. After applying the autonomous driving system to the accident-related vehicle, the tendency of intersection accidents that may occur in autonomous vehicles was identified and analyzed.

• Journal of Auto-vehicle Safety Association
• /
• v.9 no.1
• /
• pp.19-24
• /
• 2017

This paper present a performance evaluation scenarios to assess the safety performance of autonomous emergency braking (AEB) system for cyclist collision. To guarantee the safety performance of AEB for cyclist, AEB system should be tested in various scenarios which can be occurred in real driving condition. For this, real-traffic car-to-cyclist collision data are analyzed to classify the real traffic collision scenarios. Using this information, typical car-to-cyclist collision scenarios are selected. Also, in order to develop the detail features of these collision scenarios, several accident cases related with these scenarios are explained. Based on these information, test scenarios which can describe the car-to-cyclist collisions occurred in Korea are proposed. For practicality and feasibility of the test scenarios, proposed scenarios should be designed to assess the safety performance of AEB system effectively. For this, some test scenarios are combined or removed based on the consideration about the effectiveness of each scenario to the assessment of the performance of AEB system. To confirm that the proposed test scenarios are realistic and physically meaningful, simulation is conducted using simple AEB system in proposed test scenarios.

• Journal of Auto-vehicle Safety Association
• /
• v.15 no.1
• /
• pp.35-44
• /
• 2023

The purpose of this study is to predict how the actual accident changes by reconstructing the accident with an Autonomous Emergency Braking system (AEB) based on the actual accident of the LTAP-OD (Left Turn Crossing Path - Opponent Direction) intersection. A virtual AEB sensor was developed, and 150 head-on collision accident reports were secured to the insurance company to reconstruct the accident. As a result of the accident type analysis, a total of 13 types of head-on collision accidents were derived, and it is the LTAP-OD intersection accident with the highest frequency. In the LTAP-OD intersection accident, the simulation was conducted by applying the virtual AEB of each vehicle, the accident rate decreased by 90% or more when the AEB of the left-turn vehicle was applied, and the accident rate decreased by 50%. In addition, the most frequent collision types in LTAP-OD accidents were the front bumper on the driver's side of a vehicle going straight and the front bumper on the passenger's side of a vehicle turning left.

• Journal of Auto-vehicle Safety Association
• /
• v.14 no.4
• /
• pp.84-90
• /
• 2022

With the popularization of autonomous driving technology, safety has emerged as a more important criterion. However, there are no assessment protocol or methods for AES (Autonomous Emergency Steering). So, this study proposes AES assessment protocol and scenario corresponding to collision avoidance Car-to-Car scenario of Euro NCAP in order to prepare for obstacles that appear after the emergency steering of LV (Leading Vehicle) avoiding obstacles in front of. Autoware-based autonomous driving stack is developed to test and simulate scenario in CARLA. Using developed stack, it is confirmed that obstacle avoidance is successfully performed in CARLA, and the AES performance of VUT (Vehicle Under Test) is evaluated by applying the proposed assessment protocol and scenario.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.3
• /
• pp.123-129
• /
• 2014

SCC (Smart Cruise Control) and AEBS (Autonomous Emergency Braking System) are using various types of sensors data, so it is important to consider about sensor data reliability. In this paper, data from radar and vision sensor is fused by applying a Bayesian sensor fusion technique to improve the reliability of sensors data. Then, it presents a sensor fusion verification tool developed to monitor acquired sensors data and to verify sensor fusion results, efficiently. A parallel computing method was applied to reduce verification time and a series of simulation results of this method are discussed in detail.

• Journal of Auto-vehicle Safety Association
• /
• v.15 no.4
• /
• pp.23-31
• /
• 2023

In recent years, the technology for autonomous driving has been advancing rapidly, ADAS (Advanced Driver Assistance System) functions, which improve driver convenience and safety performance, are mostly equipped in recently released vehicles and range from level 0 to level 2 in autonomous driving technology. Among the various functions of ADAS, AEBS (Autonomous Emergency Braking System), which analyzes traffic accidents, is the most closely related to the vehicle's braking. This study developed a simulation technique for reproducing accidents related to AEBS based on real vehicle experimental data, and it was applied to the analysis of actual ADAS vehicle accidents to identify the causes of accidents.

• Journal of Auto-vehicle Safety Association
• /
• v.9 no.2
• /
• pp.13-19
• /
• 2017

This paper presents safety assessment scenarios for cyclist autonomous emergency braking(AEB) system. To assess the safety performance of AEB in real traffic situation with limited number of scenarios, scenario should reflect the characteristics of real traffic collision cases. For this, statistic data of real traffic car-to-cyclist collision in Korea are analyzed. Many types of accidents are listed and categorized based on the movement of vehicle and cycle just before the collision. Then, the characteristics, main issues and limitations of each scenarios are discussed. Not only the test scenario itself but also the cost and time for the test are very important issues for the test scenarios to actually repeat the test for various systems. Also, the performance of AEB can be effected by the algorithm of AEB and the technical limitation of the sensors and hardwares. Therefore, required number of tests, possibility of dummy destruction and other technical issues are discussed for each scenarios. Based on these information, typical scenarios are selected. Also, using this information, vehicle speed range, cyclist speed and collision point are established. Proposed scenarios are verified and modified based on the vehicle test results. vehicle test was evaluated 5 times for each scenarios. Based on this results, final test scenarios are modified and proposed.

• Journal of Auto-vehicle Safety Association
• /
• v.11 no.1
• /
• pp.7-16
• /
• 2019

There are trying to reduce damage from automobile accident in many countries. In many automobile companies, there have been active study on development of ADAS (Advanced Driver Assistance Systems) for commercialization, in order to reduce damage from automobile accident. ADAS is the system providing convenience and safeness for drivers. Generally, ADAS is composed of ACC (Adaptive Cruise Control), LKAS (Lane Keeping Assist System), and AEB (Autonomous Emergency Braking). AEB of the ADAS, it is an autonomous emergency braking system and it senses potential collide and avoids or degrades it. Therefore AEB plays a significant role in reducing automobile accident rate. However, AEB safety evaluation method is not established not yet. For this reason, this study suggests safety evaluation scenarios with adding cut-in, sensor malfunctioning scenario that scenario domestic street conditions considered as well as original standard AEB scenario of Euro NCAP for establishment of safety evaluation method of AEB. And verifying validity of suggested scenario by comparing the calculated values of the theoretical formulas presented in the previous study with results of the actual vehicle test.