The National Highway Traffic Safety Administration (NHTSA) and the California Department of Motor Vehicles (CA DMV) collect and utilize data from traffic accidents caused by Automated Driving Systems (ADS) driving on real roads, as a policy. Leading autonomous driving technology companies such as Tesla and Waymo collect their own driving and accident data and use them for technology advancement. ADS traffic accident data that occur when driving on real roads are valuable for identifying problems in unexpected situations. This study analyzes the risk of traffic accidents by Operational Design Domain (ODD) on ADS traffic accident data that occurred while driving on an actual road and aims to present a road traffic law-based driving ability evaluation scenario in a complex ODD configuration in high-risk situations, wherein an ADS can be particularly vulnerable in mixed traffic situations. The actual road traffic accident data of ADS from 2,289 accidents as provided by the NHTSA were analyzed. Analysis of the characteristics of ADS traffic accidents revealed that accidents occurred mainly on ordinary ODDs with high traffic demand during actual road driving, that is, on dry roads during clear days and daylight. In traffic situations including ADS and Human Driving Vehicle(HDV), approximately 40% of traffic accidents were confirmed to have occurred because of HDV colliding with stationary ADS and occurred in unexpected situations, such as changing the HDV when driving straight ahead of the ADS. Results of analyzing the risk of traffic accidents on the driving status of ADS by ODD, showed that the risk of traffic accidents that occurred while the ADS was driving straight ahead was 2.27, with dry road conditions, sunny weather, and a road speed limit of 21 to 30 mph at night when streetlights were turned on. Thus, the ADS road traffic law-based driving ability evaluation scenario can be used to evaluate whether to recognize and respond to accident risk situations by developing ADS road traffic law-based driving ability evaluation scenarios for situations vulnerable to accidents due to HDV cut-in in traffic situations that include ADS and HDV. In future, this can be used as basic data for preparing related regulations and institutional devices, such as traffic accident investigations and driving ability evaluations by ADS.

This study aimed to enhance the safety of autonomous bus services by systematically identifying safety-related factors and establishing priorities based on real-world operating environments. An expert survey was conducted using a autonomous bus currently operating in Pangyo Zero City as a case study. Building on the concept of the Operational Design Domain, a two-layer safety framework was developed consisting of four primary categories (Layer 1): physical infrastructure, operating conditions, communication environment, and weather conditions, and their corresponding detailed elements (Layer 2). A fuzzy Analytic Hierarchy Process(AHP) analysis revealed that physical infrastructure had the highest relative importance, with key safety-critical factors identified as intersection type, construction work zone, lane markings, and adverse weather. Subsequently, a strength, weakness, opportunity and threat (SWOT) analysis was employed to propose short-, mid-, and long-term strategic actions, including the enhancement of object recognition functions based on advanced camera sensor fusion, reinforcement of safety driver and onboard safety personnel systems, and establishment of infrastructure pre-notification systems for construction and maintenance activities. This study provides a quantitative prioritization of safety factors for autonomous bus services and links these findings to a practical technology and policy roadmap, contributing to the enhancement of safety and development of commercialization strategies for future autonomous public transportation services.

With the rapid transition to an aging society, the need for assistive technologies that promote independent indoor living for the elderly and mobility-impaired has become increasingly critical. This study proposes the development of a next-generation powered chair designed to support such independence by compensating for mobility limitations caused by natural aging. The proposed system incorporates two core functionalities: (1) an low seat-lifting mechanism capable of lowering the seat height to 7 cm, and (2) a short-range autonomous driving mode operable in both lowered and lifted positions. The low driving mode enables the user to approach low tables or desks and facilitates effortless transfer to and from low beds or sofas. In the lifted position, the system performs real-time obstacle detection and avoidance within a 3-meter range, preventing falls and collisions while expanding the user’s range of motion— for instance, by allowing access to higher objects or enabling eye-level communication with standing individuals. To realize these functions, a rack-and-pinion lifting mechanism is applied, along with a direct target-point designation method utilizing an LED pointer and a wiper-type screening approach for real-time obstacle avoidance. The design concept, implementation strategy, and validation plan are presented. This research contributes to enhancing the quality of life for elderly users by maximizing their remaining physical capabilities, while simultaneously reducing the physical and emotional burden on caregivers.

This study analyzes and compares the determinants of accident severity between human-driven vehicle (HDV) and autonomous vehicle (AV) mixed environments using collision data from the California Department of Motor Vehicles . To address the high dimensionality and categorical complexity of the dataset, an XGBoost-based classification model was developed and the Shapley additive explanations method was employed to explain the contribution and directional influence of each explanatory variable. An undersampling and ensemble approach was utilized to mitigate class imbalances and enhance the model stability. The results revealed that in an HDV environment, driver perception and evasive responses were dominant factors influencing crash outcomes, with collision direction and relative speed significantly affecting the severity. By contrast, in the AV–HDV mixed environment, intersection conditions and complex driving contexts were associated with higher accident severity, thus demonstrating the current limitations of AV systems in managing unstructured traffic scenarios. These findings suggest that as AV deployment progresses, the key determinants of crash severity shift from human behavioral factors to system and environmental factors, thus providing empirical insights for future AV safety evaluations and policy frameworks.

This study aims to analyze the driving trajectories and lateral behavior characteristics of autonomous vehicles via simulation and to derive the implications for roadway infrastructure design based on the analysis results. A three-lane, one-way autonomous driving simulation environment was established to replicate the actual driving characteristics of autonomous vehicles. Roadways were designed based on domestic road design standards (MLTM, 2020), where horizontal, vertical, and cross-sectional alignments were incorporated and design speeds ranging from 20 to 120 km/h were considered. Curves with minimum radii of 15, 30, 60, …, 710 m were implemented. Autonomous vehicles were driven along these designed roads to obtain driving data, including position, speed, and steering angle. The lateral deviation from the lane center was calculated for each lane by measuring the distance between the front and rear wheels of the vehicle and the lane centerline. This approach allows for the analysis of lane-specific deviation characteristics under different speeds and curve radii, thus enabling a quantitative assessment of the lateral clearance required for autonomous-vehicle operation. Lateral deviation increased when vehicles entered or exited curves, particularly in outer lanes and at curves with changing turning directions. Passenger cars and heavy vehicles showed decreasing deviations within curves, whereas the deviations varied in straight sections. The lateral clearance increased with the design speed for passenger cars, whereas heavy vehicles generally exhibited limited clearance owing to their larger size and mirror widths, with slight increases above 100 km/h. Autonomous vehicles maintained lane centers outside curve entries and exit sections, thus indicating that variable lane widths can be safely implemented. The existing design standards based on human driving may be adapted for autonomous vehicles, thus enabling more efficient roadway use while maintaining stability.

This study aims to provide a basis for selecting the appropriate traffic-flow evaluation indicators by quantitatively analyzing the relative importance of such indicators in mixed traffic environments in which automated vehicles (AVs) and conventional vehicles coexist. As AV technology progresses and its adoption increases, establishing reliable evaluation criteria that accurately reflect the characteristics and performance of traffic systems under transitional conditions is crucial. Thus, approximately 40 domestic and international studies were reviewed in this study, from which 45 evaluation indicators were identified. These indicators were classified into three major categories: mobility, safety, and environment. Five frequently used and representative indicators were selected from each category based on the appearance frequency and relevance. An analytic hierarchy process survey was conducted with a group of transportation experts to derive the relative importance (weights) of both the major categories and individual indicators. The analysis revealed that safety (0.53676) was the most important category, followed by mobility (0.34795) and environment (0.11528). After combining the weights of the categories and sub-indicators, the top three indicators, i.e., time to collision (TTC), time exposed to TTC, and deceleration rate to avoid crashes, appeared to be safety related and associated directly with the collision risk. These findings suggest that, in the early stages of AV deployment, traffic evaluations should prioritize safety considerations over mobility or environmental factors to ensure the successful integration of AVs into existing traffic systems.

운전 시뮬레이션을 이용하여 3-수준 자율주행 상황을 구현한 후 청년 및 고령운전자가 주간/야간 운전 조건과 비 운전과제(nonm-driving task: NDT) 수행 여부에 따라 보이는 제어권 인수시간(takeover time: TOT), 차량제어 (vehicle control :VC) 및 주관적 작업부하(subjective workload: SW) 수준에서의 차이를 비교하였다. 실험참가자들에 게는 자율주행 중 NDT를 수행하도록 하였고 NDT 수행 도중 제어권 인수가 요청되면 제어권을 빠르고 정확하게 인수받아 수동운전으로 전방의 장애물을 회피하도록 하였다. 본 연구 결과를 요약하면 다음과 같다. 첫째, 야간 운전 조건과 NDT 수행 조건에서 실험참가자들의 TOT는 증가하였고, 차량에 대한 종적 및 횡적수행 모두 저하되었으며, SW 수준은 더 높았다. 둘째, 청년운전자 집단에 비해 고령운전자들의 VC 수행이 상대적으로 더 저조하였다. 셋째, 고령운전자들은 야간 운전과 NDT 수행 요구가 결합되면 모든 종속측정치에서 청년운전자들에 비해 상대적으로 더 저하된 수행을 보였다. 이러한 결과는 야간 자율주행에서 고령운전자의 주의가 분산될 경우 자율주행 차량과의 상호 작용 및 긴급한 상황에서의 장애물 회피에서 어려움이 증가할 수 있다는 것을 시사한다.

The automotive industry is rapidly shifting from hardware-focused design to Software Defined Vehicles (SDVs), where functions are flexibly updated through software. Embedded systems are central to this transition, ensuring real-time data processing and control across sensors, actuators, and controllers. Yet, most autonomous driving education and competitions have been designed for senior students, creating high entry barriers for early undergraduates. This study proposes an embedded practice-based education model for lower-year students, implemented through an autonomous driving competition. Arduino was adopted as an accessible embedded platform, enabling rapid prototyping and intuitive learning of sensor–controller–actuator integration. The curriculum was structured to advance from interrupt-based programming to Real-Time Operating System (RTOS)-based task scheduling, providing stepwise exposure to core SDV concepts. The model was validated through a mission-oriented competition that included line following, obstacle avoidance, and stop-line detection tasks. Dual assessment—combining technical performance indicators with rubric-based educational outcomes— demonstrated both algorithmic feasibility and pedagogical effectiveness. This work highlights that early undergraduates can gain meaningful SDV-oriented embedded control experience through lightweight competitions. The proposed framework offers an effective pathway for cultivating the next-generation mobility workforce, bridging the gap between theoretical education and practical implementation in the SDV era.

In this study, we comparatively analyzed the efficiency of conventional image recognition methods and propose a digital information provisioning method for autonomous vehicle traffic safety facility recognition. We evaluated the practicality of both approaches from the perspective of autonomous vehicles' capabilities of processing regulatory information and the distribution of legal responsibility. Comprehensive field experiments were conducted at 9 major intersections in the Pangyo Techno Valley area of Hwaseong City over a 10- day period from July 12-23, 2021. Three test vehicles equipped with in-vehicle terminals and video cameras collected data through 300 driving scenarios, including 240 during peak hours and 60 during off-peak periods. The proposed digital information provision method exhibited superior performance, achieving a 100.0 % recognition success rate across all test scenarios and road conditions. In contrast, the conventional image recognition method exhibited significant variability in performance, ranging from 56.9 % in underpass conditions to 95.9 % in areas with communication interference, with an overall average of 70.8 %. The digital information provision method demonstrated superior performance compared to conventional image recognition approaches for autonomous vehicle regulatory compliance. The proposed approach delivered consistent and reliable information regardless of physical obstacles or environmental conditions. This method ensures complete comprehension of regulatory information, which is essential for establishing clear legal responsibility frameworks in autonomous driving environments.

In this study, we propose a data-driven analytical framework for systematically analyzing the driving patterns of autonomous buses and quantitatively identifying risky driving behaviors at the road-segment level using operational data from real roads. The analysis was based on Basic Safety Message (BSM) data collected over 125 days from two Panta-G autonomous buses operating in the Pangyo Autonomous Driving Testbed. Key driving indicators included speed, acceleration, yaw rate, and elevation, which were mapped onto high-definition (HD) road maps. A hybrid clustering method combining self-organizing map (SOM) and k-means++ was applied, which resulted in eight distinct driving pattern clusters. Among these, four clusters exhibited characteristics associated with risky driving such as sudden acceleration, deceleration, and abrupt steering, and were spatially visualized using digital maps. These visualizations offer practical insights for real-time monitoring and localized risk assessment in autonomous vehicle operations. The proposed framework provides empirical evidence for evaluating the operational safety and reliability of autonomous buses based on repeated behavioral patterns. Its adaptability to diverse urban environments highlights its utility for intelligent traffic control systems and future mobility policy planning.

In this study, we investigated and analyzed the impact of changes in driving speed and inter-vehicle distance on users’ perceived tension during autonomous vehicle operation. To this end, a survey experiment was conducted for both urban roads and highways. The results show that the greatest changes in perceived tension occurred in the range of 50–70 Km/h and 50–70 m following distance on urban roads, and in the range of 80–100 Km/he and 60–80 m following distance on highways. Furthermore, modeling user behavioral responses to perceived tension based on changes in speed and following distance revealed that linear models best described the relationship for speed on both urban roads and highways. For the following distance, a quadratic model was the most suitable for urban roads, whereas a logarithmic model best fit the highway data. These findings are expected to contribute to practical operational guidelines for autonomous vehicles by alleviating users’ psychological discomfort and enhancing public acceptance. Future research will extend this study using a driving simulator to examine user responses in more realistic driving environments.

In the Autonomous Mobility Living Lab, traffic situations with both autonomous vehicles (AVs) and ordinary vehicles driven by humans (HDVs) are explored. Research on countermeasures and efficient transportation management plans has emerged from this context. In this study, we analyzed the effect of AVs with different speeds on signal intersections and road networks to derive efficient traffic operation plans for roads on which various AVs and HDVs with different driving behaviors are mixed in Living Lab cities. To that end, we conducted a simulation-based analysis of the effects of AV mixing rates on continuous signal intersections and the road network in traffic situations where AVs and HDVs were mixed at peak and non-peak driving hours. The simulation scenario was designed by classifying the traffic volume levels at peak and non-peak times and defining various AV mixing rates; we also set the driver behaviors of the AVs as either conservative or aggressive. By performing a small-scale traffic simulation, the average control delay, average stopped delay, average queue length, and average travel time of the signal intersection for each scenario were derived, and the impact of the AV mixing rate on traffic operation was analyzed. The results of the analysis show that higher AV mixing rates were associated with lower measurements of the effectiveness of signal intersections, which had a positive effect on traffic operation. This resulted in a stable and efficient improvement of the traffic flow at intersections as more vehicles passed through at the time of the allocated signal, as the AVs in the simulation could be driven at short intervehicle intervals by receiving real-time traffic information. In the traffic operation on the network, we found that the higher the AV mixing rate, the lower the average travel time, resulting in a greater effect of facilitating the traffic flow of the urban network. These simulated results indicate that higher AV mixing rates were associated with positive outcomes in terms of signal intersections and network traffic operation. We expect that this simulation can be used to establish real traffic operation plans in traffic situations where AVs are mixed at each stage of autonomous driving technology in the future.

Autonomous vehicles are widely expected to be commercialized in the near future. This would naturally lead to situations in which existing vehicles and autonomous vehicles would be on the road at the same time, which would pose a notable hazard to traffic safety. From this perspective, high-risk factors relating to this deployment should be identified to prepare measures to promote traffic safety. However, at this point, deriving high-risk factors based on actual data is problematic because autonomous vehicles have not yet been widely commercialized. In this study, we derive high-risk factors that would apply if autonomous vehicles were allowed to drive alongside vehicles driven by humans using a meta-analysis. We synthesized factors related to autonomous vehicles mentioned in the relevant literature. An analysis was conducted based on a total of 58 documents according to five keywords related to autonomous vehicles (crash factors, scenarios, predictive models, laws, and regulations). We also performed a binary meta-analysis of factors related to autonomous vehicles according to these keywords and a meta-analysis of effect size according to the relative size of factors to evaluate them comprehensively. We found that many different aspects of driving such as navigating intersections, lanes, fog, rain, acceleration and deceleration, rear-end collisions, inter-vehicle spacing, and pedestrian collisions were notable as high-risk factors. This study provides basic data to identify high-risk factors to support the development of related prediction models.

본 연구는 레벨 3 자율주행의 운전이양권(TOR) 안전성 향상을 위해, 기존 행동 기반 감지 방식의 한계를 극복하 는 운전자 모니터링 시스템(DMS)을 개발했다. 차량의 미러 내장형 RGBW 카메라를 이용한 비접촉 원격 광용적맥 파(rPPG) 기술로 운전자의 심박수를 실시간 측정하고, 심박변이도(HRV) 분석을 통해 졸음, 스트레스 등 운전자의 각성 수준을 판단한다. 딥러닝 기반 얼굴 인식, 신호 처리, 패턴 인식 알고리즘을 통합하여 시스템을 구현했다. 총 28명을 대상으로 105시간 이상의 실제 도로 환경에서 검증한 결과, 심전도(ECG) 대비 85.14%의 심박수 측정 정확 도와 90.81%의 상태 판단 정확도를 달성했다. 본 연구는 생체신호 기반의 운전자 상태 평가가 TOR 판단의 신뢰성 을 높이는 핵심 기술이 될 수 있음을 실증했다.

이동로봇은 인공지능, 센서 기술 등과 융합함으로서 다양한 산업 및 서비스 분야에서 광범위하게 사용되고 있으며, 조선 및 해 양 분야에서도 이동로봇을 활용한 물품 운반, 현장 모니터링, 위험한 업무 등에 대한 연구가 수행됨으로서 생산성 향상 및 안정성 강화를 향상시키고자 하고 있다. 본 연구에서는 선박기관실처럼 내연기관, 선반, 드릴머신, 공구대, 용접실습대 등 다양한 기기 및 장비의 간격이 좁고 구조가 복잡한 환경의 기관실습실 내에서 이동로봇의 자율주행을 구현함으로서 선박기관실에 적용 가능여부를 확인하고자 하였다. ROS2기반의 이동로봇으로 SLAM 라이브러리 중 하나인 Cartographer를 사용하여 지도를 작성한 후 여러 위치에서 자율주행 시험과 지도에 없는 장애물을 놓은 경우 자율주행 시험결과 복잡한 환경에서도 높은 자율주행 성능을 확인하였다. 선박기관실은 실험한 장소와 여러환 경의 차이는 있으나 구조의 변화가 거의 없어 자율주행이 가능할 것으로 사료된다.

Autonomous vehicle (AV) technology is rapidly entering the commercialization phase driven by advancements in artificial intelligence, sensor fusion, and communication-based vehicle control systems. Real-world road testing and pilot deployments are increasingly being conducted, both domestically and internationally. However, ensuring the safe operation of AVs on public roads requires not only technological advancement of the vehicle itself but also a thorough pre-evaluation of the road environments in which AVs are expected to operate. However, most previous studies have focused primarily on improving internal algorithms or sensor performance, with relatively limited efforts to quantitatively assess how the structural and physical characteristics of road environments affect AV driving safety. To address this gap, this study quantitatively evaluated the compatibility of road environments for AV operation and defined the road conditions under which AVs can drive safely. Three evaluation scenarios were designed by combining static factors such as curve radius and longitudinal gradient with dynamic factors such as level of service (LOS). Using the MORAI SIM autonomous driving simulator, we modeled the driving behaviors of autonomous vehicles and buses in a virtual environment. For each scenario, the minimum time to collision (mTTC) from the moment the AV sensors detected a lead vehicle was calculated to assess risk levels across different road conditions.The analysis revealed that sharper curves and lower service levels resulted in significantly increased risk. Autonomous buses exhibited a higher risk on downhill segments, autonomous vehicles were more vulnerable to uphill slopes and gradient transitions. The findings of this study can be applied to establish road design standards, develop pre-assessment systems for AV road compatibility, and improve AV route planning and navigation systems, thereby providing valuable implications for policy and infrastructure development.