Recent, the issue of the fourth industrial revolution triggered by technological advances has changed the automobile industry centered on internal combustion engines, and quantitative growth of the global automobile market, which has grown rapidly, has been slowing since 2015. These advances in technology are expected to develop beyond the advanced driver assistance system to autonomous driving technology. According to SAE-J3016 published by the Society of Automotive Engineers, the technology of autonomous vehicles is divided into a total of six stages according to the driver's intervention and automation level from 0 to 5. Securing safety for autonomous vehicles is important. But, research on safety evaluation theory and autonomous vehicle evaluation method based on real vehicle test is insufficient. In this study, the longitudinal distance theory equation and continuous test scenario were proposed for the test method of autonomous vehicles for fixed targets, and the real vehicle test was conducted. When comparing the theoretical values compared to the measured values, it was determined that it was reliable with a minimum error rate of 0.484% and a maximum error rate of 7.391%. Using the proposed theoretical equation, it is judged that it can be used as a safety evaluation method in an environment where real vehicle test is not possible because it can grasp the trend in the longitudinal direction in the development stage.

In this paper, we propose the improved EDR (Event Data Recorder) storage data, which can identify the cause of unintended acceleration of eco-friendly vehicles. The proposed EDR storage data includes the brake pressure sensor value and a brake pedal travel sensor value. To verify the proposed EDR storage data, we observe the control algorithm and internal structure of the vehicle dynamic control system and a regenerative braking system in an eco-friendly vehicle.

Among the various functions of ADAS (Advanced Driver Assistance System), the most important and representative function to the safety of vehicle passengers is AEB (Autonomous Emergency Braking system). In South Korea, laws are in progress from 2022 for making it mandatory for passenger vehicles to be installed. And as AEB-equipped vehicles continues to increase in the future, the demand for accident analysis related to the AEB function is expected to increase in the future. In order to find out the operating limits of AEB, it is necessary to consider the situations exceeding the standards covered by EuroNCAP. Therefore we have performed four experiments in this study, including situations encountered in real-word traffic conditions, i.e., an oblique stop of Global Vehicle Target (GVT) and ADAS sensor failures. These experimental results are expected to be of great help in accurate and reliable accident analysis by considering them when analyzing traffic accidents for ADAS vehicles.

The CFT(Cornering Fatigue test) and RFT(Radial Fatigue Test) are tests for evaluating the endurance of the disc and rim region of the wheel. In recent, automobile wheel safety standards have been revised and the applied load and target life criteria are different from existing conditions. The verification evaluation of all wheels requires a lot of time and cost. In this study, the endurance severity of each test was compared through strain-life approach by selecting 4 steel and 8 aluminum wheels.

In this study, the driving performances of elderly who's age is over 65 were evaluated. The driving simulation was conducted using a compact driving simulation (CDS) and the simulation scenarios were developed from actual roads by replicating geometry of skewed intersection and traffic control devices located in Jungnang-gu, Seoul, Korea. 27 elderly drivers and 10 non-elderly drivers were recruited and participated on the virtual turning right and going straight driving experiment of CDS. Virtual driving data of driving time, speed, distance, acceleration and deceleration speeds, brake power, and steering wheel rotation angle were recorded and analyzed. Generally, elderly driver took more times to pass through the skewed intersection road and showed lower approaching speed as much as 40% and 25% in case of turning right and going straight scenarios respectively. The speed deviation at skewed intersection road between elderly and non-elderly driver is expected to cause frequent lane changes and overtaking.

Composite material is recently very important material for eco-friendly vehicles because of its excellent mechanical property and lightweight effect. So, many research results have been recently published for developing the composite material to apply vehicles. In this paper, new brake system is presented using composite material to response this situation. And advantages in terms of performance compared to competitive company will be discussed in depth to verify superiorities of the new composite brake. To do so, composite brake systems which have the same size as the competitive company to the same vehicle is applied. And superiorities through a variety of test results are presented. First, normal braking performances are compared with competitive company through braking effect, heat capacity and friction test, Second, circuit driving and high speed fade test are also verified with competitive company to confirm harsh braking performances for the new composite brake system. Finally, the effects of applying the composite brake to automobile industry like electric car are analyzed.

A V2X system can be a candidate as a means to increase the stability of autonomous vehicles. In particular, in order to implement a Level 4 or higher autonomous driving system, the application of the V2X system is essential. Wireless communication technologies applicable to the V2X system include WAVE and C-V2X. Currently, the V2X service most used by autonomous driving systems is a service that provides signal phase and timing information and since real-time characteristic is a very important, verification of this service must be done. In this paper, we measured the time delay characteristics for providing signal phase and timing information using WAVE and LTE communication, and proposed a TOD-based signal phase and timing information generation method without using V2X communication system. To analyze the time delay characteristics, RTT (Round Trip Time) was measured as a result of the measurement. Average RTT using WAVE communication was 5.84ms and was 104.15ms with LTE communication. As a result of measuring the error between the signal phase and timing information generated based on TOD and the actual traffic light state, it was measured to be -0.284~3.784sec.

In the automotive industry, the development of automobiles to meet safety requirements is becoming increasingly complex. This is because quality evaluation agencies in each country are continually strengthening new safety standards for vehicles. Among these various requirements, collision safety must be satisfied by controlling airbags, seat belts, etc., and can be defined as post-crash safety. Apart from this safety system, the Advanced Driver Assistance Systems (ADAS) use advanced detection sensors, GPS, communication, and video equipment to detect the hazard and notify driver before the collision. However, research to improve passenger safety in case of an accident by using the sensor of active safety represented by ADAS in the existing passive safety is limited to the level that utilizes the sudden braking level of the FCA (Forward Collision-avoidance Assist) system. Therefore, this study aims to develop logic that can improve passenger protection in case of an accident by using ADAS information and driving information secured before a collision. The proposed logic was constructed based on LSTM deep learning techniques and trained using crash test data.

Lane change in urban environments is a challenge for both human-driving and automated driving due to their complexity and non-linearity. With the recent development of deep-learning, the use of the RNN network, which uses time series data, has become the mainstream in this field. Many researches using RNN show high accuracy in highway environments, but still do not for urban environments where the surrounding situation is complex and rapidly changing. Therefore, this paper proposes a lane change possibility decision network by adopting Attention layer, which is an SOTA in the field of seq2seq. By weighting each time step within a given time horizon, the context of the road situation is more human-like. A total 7D vectors of x, y distances and longitudinal relative speed of side front and rear vehicles, and longitudinal speed of ego vehicle were used as input. A total 5,614 expert data of 4,098 yield cases and 1,516 non-yield cases were used for training, and the performance of this network was tested through 1,817 data. Our network achieves 99.641% of test accuracy, which is about 4% higher than a network using only LSTM in an urban environment. Furthermore, it shows robust behavior to false-positive or true-negative objects.

In autonomous driving cars, 3D pose estimation can be one of the effective methods to enhance safety control for OOP (Out of Position) passengers. There have been many studies on human pose estimation using a camera. Previous methods, however, have limitations in automotive applications. Due to unexplainable failures, CNN methods are unreliable, and other methods perform poorly. This paper proposes robust real-time multi-human 3D pose estimation architecture in vehicle using monocular RGB camera. Using particle filter, our approach integrates CNN 2D/3D pose measurements with available information in vehicle. Computer simulations were performed to confirm the accuracy and robustness of the proposed algorithm.

The Korea's automobile safety standards consist of standards for automobiles, automated vehicles, two-wheeled vehicles and auto parts, and there are 157 articles. In connection with the safety standards, the implementation rules stipulated for detailed test methods and procedures are composed of the main body, asterisks and annexes, and have a rather complicated structure. In addition, the test items in the asterisk are specified with the same or similar name as the test items in other asterisks. In this study, the structure of these implementation rules is simplified and the same test items are integrated. This is expected to increase the user's understanding of laws and regulations and efficiency.