Search Results (759)

This study analyzed the impact of human drivers’ Keep Right Rule noncompliance on sustainable freeway operations in mixed traffic. Using the microscopic traffic simulation tool, a total of 36 scenarios were examined based on variations in driving behavior, presence of slow vehicles in the passing lane, desired speed, and number of lanes. The Wiedemann-99 car-following model and autonomous driving logic were applied for simulation. Simulation results revealed that the occupation of the passing lane by a human-driven slow vehicle increased the recovery time and variability in right-side rule compared to free lane selection. Also, 20 km/h was a threshold desired speed gap that activated the bottleneck by the slow vehicle in a passing lane. Lastly, as the number of lanes increased, bottleneck formation was diminished. The findings point to a mixed traffic systemic paradox. Human drivers can alleviate bottleneck formation by flexibly performing right-side overtaking even though it is illegal, whereas autonomous vehicles cannot perform right-side overtaking, which unintentionally activates a bottleneck under strict rule compliance. These results show that in mixed traffic conditions, even minor violations of traffic rules by human drivers can lead to congestion. Therefore, to achieve sustainable and safe road traffic by harmonizing mixed traffic, institutional improvements are necessary alongside advances in autonomous driving technology. Full article

Compared to cameras and LiDAR, radar provides superior robustness under adverse conditions, as well as extended sensing range and inherent velocity measurement, making it critical for surveillance applications. To advance research in deep learning-based radar perception technology, several radar datasets have been publicly released. However, most of these datasets are designed for autonomous driving applications, and existing radar surveillance datasets suffer from limited scene and target diversity. To address this gap, we introduce KuRALS, a range–Doppler (RD)-level radar surveillance dataset designed for learning-based long-range detection of moving targets. The dataset covers aerial (unmanned aerial vehicles), land (pedestrians and cars) and maritime (boats) scenarios. KuRALS is real-measured by two Kurz-under (Ku) band radars and contains two subsets (KuRALS-CW and KuRALS-PD). It consists of RD spectrograms with pixel-wise annotations of categories, velocity and range coordinates, and the azimuth and elevation angles are also provided. To benchmark performance, we develop a lightweight radar semantic segmentation (RSS) baseline model and further investigate various perception modules within this framework. In addition, we propose a novel interference-suppression loss function to enhance robustness against background interference. Extensive experimental results demonstrate that our proposed solution significantly outperforms existing approaches, with improvements of $10.0\%$ in mIoU on the KuRALS-CW dataset and $9.4\%$ on the KuRALS-PD dataset. Full article

In a fully autonomous vehicle, the driver becomes a passenger, free to adopt different seating positions. This change challenges traditional passive safety systems—such as seatbelts, airbags and seat design—that are optimised for a forward-facing position. As autonomous vehicles are integrated into mixed traffic with conventional cars, solutions need to address these challenges. In this intermediate stage, fully autonomous cars will need a system that, in the event of an accident, can rotate the seats to the most ideal position tested by the manufacturer. This could be a number of positions where the seat, airbags and seatbelts are optimised, taking into account the expected direction of impact. It is important that the rotation is not too radical, as this would increase the risk of injury. In addition, the seat dimensions need to be increased to improve energy absorption in the event of a collision, thereby reducing the impact forces on the occupants and improving overall safety. To improve passive protection, airbags will continue to be used in the future, but in completely new positions, sizes and shapes. This research aims to identify potential passive occupant safety solutions for seat positions that have been rotated in fully autonomous vehicles. The finite element simulation model on which the results in this article are based was developed in an earlier phase of the research. The current research combines two previously conducted research directions, using the modified seat and the developed airbag concept. This research’s main outcome is a system that effectively protects occupants in rotated seat positions. It maintains all evaluated injury criteria below their threshold limits and ensures controlled occupant kinematics. Full article

The automatic emergency braking (AEB) system plays a crucial role in reducing rear-end collisions and is mandatory on certain heavy-duty vehicles, with future regulations extending to passenger cars. However, most current AEB systems are designed based on onboard sensors such as cameras and radar, which may fail to prevent collisions in scenarios where the lead vehicle is already in a collision. To address this issue, this study proposes an enhanced AEB control method based on Vehicle-to-Vehicle (V2V) communication and onboard sensors. The method utilizes V2V communication and onboard sensors to predict obstacles ahead, applying effective braking when necessary. Simulation results in Matlab/Simulink R2022a show that the proposed V2V-based AEB control method reduces the risk of chain collisions, ensuring that the ego vehicle can avoid rear-end collisions even when the lead vehicle is involved in a crash. Three simulation scenarios were designed, where both the subject vehicle and the lead vehicle travel at 120 km/h. The following three distances between the subject vehicle and the lead vehicle were considered: 45 m, 70 m, and 30 m. When the lead vehicle detects an obstacle 30 m ahead and suddenly applies emergency braking, the lead vehicle fails to avoid a collision. In this case, the subject vehicle, equipped only with onboard sensors, is also unable to successfully avoid the crash. However, when the subject vehicle is equipped with both onboard sensors and vehicle-to-vehicle communication, it can prevent a rear-end collision with the lead vehicle, maintaining a vehicle-to-vehicle distance of 1 m, 6.8 m, and 3.1 m, respectively, during the stopping process. This control method contributes to advancing the active safety technologies of autonomous vehicles. Full article

Urban tunnel portals constitute critical safety zones for autonomous vehicles, where abrupt luminance transitions, shortened sight distances, and densely distributed structural and traffic elements pose considerable challenges to perception reliability. Existing driving scenario datasets are rarely tailored to tunnel environments and have not quantitatively evaluated how specific infrastructure components influence perception latency in autonomous systems. This study develops a requirement-driven framework for the identification and sensitivity ranking of information perception elements within urban tunnel portals. Based on expert evaluations and a combined function–safety scoring system, nine key elements—including road surfaces, tunnel portals, lane markings, and vehicles—were identified as perception-critical. A “mandatory–optional” combination rule was then applied to generate 48 logical scene types, and 376 images after brightness (30–220 px), blur (Laplacian variance ≥ 100), and occlusion filtering (≤0.5% pixel error) were obtained after luminance and occlusion screening. A ResNet50–PSPNet convolutional neural network was trained to perform pixel-level segmentation, with inference rate adopted as a quantitative proxy for perceptual sensitivity. Field experiments across ten urban tunnels in China indicate that the model consistently recognized road surfaces, lane markings, cars, and motorcycles with the shortest inference times (<6.5 ms), whereas portal structures and vegetation required longer recognition times (>7.5 ms). This sensitivity ranking is statistically stable under clear, daytime conditions (p < 0.01). The findings provide engineering insights for optimizing tunnel lighting design, signage placement, and V2X configuration, and offers a pilot dataset to support perception-oriented design and evaluation of urban tunnel portals in semi-enclosed environments. Unlike generic segmentation datasets, this study quantifies element-specific CNN latency at tunnel portals for the first time. Full article

The emergence of self-driving vehicles raises not only technological challenges, but also profound moral and legal challenges, especially when the decisions made by these vehicles can affect human lives. The aim of this study is to examine the moral and legal dimensions of algorithmic decision-making and their codifiability, approaching the issue from the perspective of the classic trolley dilemma and the principle of double effect. Using a normative-analytical method, it explores the moral models behind decision-making algorithms, the possibilities and limitations of legal regulation, and the technological and ethical dilemmas of artificial intelligence development. One of the main theses of the study is that in the case of self-driving cars, the programming of moral decisions is not merely a theoretical problem, but also a question requiring legal and social legitimacy. The analysis concludes that, given the nature of this borderline area between law and ethics, it is not always possible to avoid such dilemmas, and therefore it is necessary to develop a public, collective, principle-based normative framework that establishes the social acceptability of algorithmic decision-making. Full article

In recent years, artificial intelligence technology has developed rapidly, and the automobile industry has launched autonomous driving systems. However, autonomous driving systems installed in unmanned vehicles still have room to be strengthened in terms of cybersecurity. Many potential attacks may lead to traffic accidents and expose passengers to danger. We explored two potential attacks against autonomous driving systems: stroboscopic attacks and colored light illumination attacks, and analyzed the impact of these attacks on the accuracy of traffic sign recognition based on deep learning models, such as convolutional neural networks (CNNs) and You Only Look Once (YOLO)v5. We used the German Traffic Sign Recognition Benchmark dataset to train CNN and YOLOv5 to establish a machine learning model, and then conducted various attacks on traffic signs, including the following: LED strobe, various colors of LED light illumination and other attacks. By setting up an experimental environment, we tested how LED lights with different flashing frequencies and light color changes affect the recognition accuracy of the machine learning model. From the experimental results, we found that, compared to YOLOv5, CNN has better resilience in resisting the above attacks. In addition, different attack methods will interfere with the original machine learning model to some extent, affecting the ability of self-driving cars to recognize traffic signs. This may cause the self-driving system to fail to detect the presence of traffic signs, or make incorrect decisions about identification results. Full article

The swift advancement of autonomous driving technology in China renders the traditional driver-centred liability framework inadequate for the regulatory demands of advanced automation. Traffic accidents involving advanced autonomous cars frequently provide difficulties in identifying responsible parties and assigning liability. This study employs a comparative analytical approach to evaluate the liability regimes utilised across different jurisdictions, such as the driver liability, the system liability, the manufacturer and operator liability, and the composite liability regimes. It proposes that liability standards ought to differ according to levels of automation, mirroring the benefits and constraints of each regime within China’s legal and industrial framework. Liability should be assigned to the driver at Levels 0–2, divided between the driver and manufacturer or operator at Level 3, contingent upon road and system circumstances, and predominantly attributed to manufacturers, operators, and system providers at Levels 4–5. This study outlines a framework for enhancing China’s autonomous vehicle liability system and aligning legal accountability with technological advancements, while offering recommendations for other jurisdictions in regulating developing technology. Full article

With the widespread adoption of camera and LiDAR sensors, 3D multi-object tracking (MOT) technology has been extensively applied across numerous fields such as robotics, autonomous driving, and surveillance. However, existing 3D MOT methods still face significant challenges in addressing issues such as false detections, ghost trajectories, incorrect associations, and identity switches. To address these challenges, we propose a lightweight 3D multi-object tracking framework via collaborative camera and LiDAR sensors. Firstly, we design a confidence inverse normalization guided ghost trajectories suppression module (CIGTS). This module suppresses false detections and ghost trajectories at their source using inverse normalization and a virtual trajectory survival frame strategy. Secondly, an adaptive matching space-driven lightweight association module (AMSLA) is proposed. By discarding global association strategies, this module improves association efficiency and accuracy using low-cost decision factors. Finally, a multi-factor collaborative perception-based intelligent trajectory management module (MFCTM) is constructed. This module enables accurate retention or deletion decisions for unmatched trajectories, thereby reducing computational overhead and the risk of identity mismatches. Extensive experiments on the KITTI dataset show that the proposed method outperforms state-of-the-art methods across multiple performance metrics, achieving Higher Order Tracking Accuracy (HOTA) scores of 80.13% and 53.24% for the Car and Pedestrian categories, respectively. Full article

Reliable transfer of control policies from simulation to real-world robotic systems remains a central challenge in robotics, particularly for car-like mobile robots. Digital Twin (DT) technology provides a robust framework for high-fidelity replication of physical platforms and bi-directional synchronization between virtual and real environments. In this study, a DT-based testbed is developed to train and evaluate an imitation learning (IL) control framework in which a neural network policy learns to replicate the behavior of a hybrid Model Predictive Control (MPC)–Backstepping expert controller. The DT framework ensures consistent benchmarking between simulated and physical execution, supporting a structured and safe process for policy validation and deployment. Experimental analysis demonstrates that the learned policy effectively reproduces expert behavior, achieving bounded trajectory-tracking errors and stable performance across simulation and real-world tests. The results confirm that DT-enabled IL provides a viable pathway for Sim2Real transfer, accelerating controller development and deployment in autonomous mobile robotics. Full article

Runout is a common failure of brake discs. The detection of this fault usually depends on the driver, as there is a vibration in the car and on the brake pedal. As Advanced Driver Assistant Systems are implemented and autonomous driving modes are available, braking is carried out by the car instead. Brake disc runout can cause longer braking distance, so it is essential to recognize and repair it. NVH measurements have been validated to be one of the solutions to detect the fault immediately without disassembling the brake unit. In this article, the previous vibration measurements are extended with other methods that can also be used for fault detection. Brake fluid pressure measurement and integration of the disc rotation angle sensor enable the detection of faults without additional sensors. The aim of the research is to design a measurement method that can be compared with previously validated measurements. Full article

Autonomous vehicles are often subjected to disturbances that compromise path-tracking accuracy and stability. Traditional chassis controllers that rely on fixed vehicle models exhibit performance limitations under such uncertainties. To address this challenge, we propose an adaptive integrated chassis control strategy that combines a linear quadratic regulator (LQR) and a model reference adaptive control (MRAC) framework. The LQR component generates nominal control commands, while the MRAC framework compensates in real time for model uncertainties and external disturbances. Simulation studies conducted in CarMaker and MATLAB/Simulink indicate that the proposed controller substantially improves path-tracking performance. Compared with conventional methods, the proposed controller reduces the root mean square error, peak error, and integral of the absolute error by up to 25.2%, 33.5%, and 34.6%, respectively. Overall, the proposed adaptive chassis controller shows enhanced vehicle robustness and stability in simulation under challenging driving conditions. Full article

This study presents the development and implementation of an electronically actuated steering system in a Formula Student Driverless race car, aiming to support autonomous driving capability. A DC motor with a belt-drive mechanism was integrated into the original steering rack assembly without altering its core mechanical characteristics. The research also includes a validation of the steering geometry using both physical measurements and CAD simulations. The objective of this measurement is to determine the steering angle as a function of the steering wheel input angle, ensuring that the resulting data accurately informs vehicle dynamics models such as the kinematic bicycle model. These steps form the basis for closed-loop control integration in the autonomous driving platform. Full article

In this paper, a time domain and a spectral analysis of the lateral dynamics of a Lexus passenger car are presented. Measurements were made of the vehicle’s lateral acceleration and steering angle. The aim of the measurements is to understand the vehicle’s lateral dynamics during different cornering maneuvers. For this purpose, part of the measurements is performed with a driver and the other part with autonomous control. The data processed and analyzed in this research can be used to determine the nature of the lateral dynamics, which is essential to establishing a mathematical relationship between the measured signals. This will allow the identification and modeling of vehicle dynamics, which is key to the development and optimization of autonomous vehicle control systems. Full article

As the development of autonomous vehicles (AVs) progresses, new seating arrangements are emerging. Face-to-face seating is common in SAE L4 AV concepts and is intended to facilitate social interaction during autonomous driving, enabling previously unfeasible non-driving related tasks (NDRTs). However, this is countered by the unpopularity of rearward seating orientations, which is particularly pronounced in cars. In order to develop countermeasures to address this unpopularity, a deeper understanding of the underlying mechanisms is required. This study validates a model that predicts the acceptance of AVs and takes seating orientation into account. To this end, a study with N = 46 participants was conducted to investigate the influence of seating orientation on AV acceptance and related factors such as transparency, trust, and motion sickness. Additionally, internal human–machine interfaces (iHMIs) were evaluated in regard to their ability to compensate for the disadvantages of a rearward seating orientation. To achieve a realistic implementation of a fully functional SAE L4 AV, an experimental vehicle was equipped with a steering and pedal robot, performing self-driven journeys on a test track. The iHMIs provided information about upcoming maneuvers and detected road users. While engaged in a social NDRT, participants experienced a total of six journeys. Seating orientation and iHMI visualization were manipulated between journeys. Rearward-facing passengers showed lower levels of trust and higher levels of motion sickness than forward-facing passengers. However, the iHMIs had no effect on acceptance or related factors. Based on these findings, an updated version of the model is proposed, showing that rearward-facing passengers in autonomous vehicles pose a particular challenge for trust calibration and motion sickness mitigation. During NDRTs, iHMIs which depend on the attention of AV occupants for information transfer appear to be ineffective. Implications for future research and design of iHMIs to address this challenge are discussed. Full article