Driver Assistance Systems Research Articles

Knowing me, knowing you-A study on top-down requirements for compensatory scanning in drivers with homonymous visual field loss.

It is currently still unknown why some drivers with visual field loss can compensate well for their visual impairment while others adopt ineffective strategies. This paper contributes to the methodological investigation of the associated top-down mechanisms and aims at validating a theoretical model on the requirements for successful compensation among drivers with homonymous visual field loss. A driving simulator study was conducted with eight participants with homonymous visual field loss and eight participants with normal vision. Participants drove through an urban surrounding and experienced a baseline scenario and scenarios with visual precursors indicating increased likelihoods of crossing hazards. Novel measures for the assessment of the mental model of their visual abilities, the mental model of the driving scene and the perceived attention demand were developed and used to investigate the top-down mechanisms behind attention allocation and hazard avoidance. Participants with an overestimation of their visual field size tended to prioritize their seeing side over their blind side both in subjective and objective measures. The mental model of the driving scene showed close relations to the subjective and actual attention allocation. While participants with homonymous visual field loss were less anticipatory in their usage of the visual precursors and showed poorer performances compared to participants with normal vision, the results indicate a stronger reliance on top-down mechanism for drivers with visual impairments. A subjective focus on the seeing side or on near peripheries more frequently led to bad performances in terms of collisions with crossing cyclists. The study yielded promising indicators for the potential of novel measures to elucidate top-down mechanisms in drivers with homonymous visual field loss. Furthermore, the results largely support the model of requirements for successful compensatory scanning. The findings highlight the importance of individualized interventions and driver assistance systems tailored to address these mechanisms.

Traffic Sign Recognition and Classification using Deep Neural Networks

Deep Neural Networks such as Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), have been applied in various applications, including machine vision and computer vision. One of the most notable emerging applications of deep neural networks is Advanced Driver Assistance System (ADAS). In ADAS model, the deep neural networks have outperformed the manual human performance in terms of traffic sign recognition and classification. to the problem of traffic sign recognition, with promising results. This study has proposed a novel Convolutional Neural Network (CNN) architecture for recognizing and classifying traffic signs. The German Traffic Sign Recognition Benchmark dataset is used for experimental analysis. The outcome shows that the proposed model performs comparatively better. The results also indicate that the implemented CNN model with Adam optimizer and RankMax activation performs well for recognizing and classifying traffic sign images.

Perceived rain dynamics on hydrophilic/hydrophobic lens surfaces and their influences on vehicle camera performance

Cameras are increasingly used in modern vehicles equipped with advanced driver assistance systems (ADAS) to collect environmental information. Cameras suffer performance degradation when driving in adverse weather conditions, such as rain, as precipitation droplets impact the camera lens and cause obstruction and blurring of the vision. The relationships between image quality, object detection accuracy, and surface wettability of camera lenses are investigated. This paper applies a previously developed evaluation procedure for wind tunnel testing with simulated adverse driving and rain conditions. Realistic rain characteristics perceived by a moving vehicle at different driving speeds are simulated using a novel rain simulation system implemented into a wind tunnel. Moreover, an emphasis is put on comparing the use of hydrophilic and hydrophobic surfaces to provide insights into material selection when designing camera lenses for ADAS. It is found that droplet dynamics, such as size, velocity, shape, and motion can impact the camera image quality and, subsequently, object detection accuracy. This paper demonstrates the use of various materials and evaluation metrics and their implications from a practical perspective when subjected to realistic driving-in-rain scenarios. The results suggest that the use of hydrophobic lenses promotes better performance over hydrophilic lenses in most cases with exceptions.

Integration of Computer Vision and IOT Into an Automatic Driving Assistance System for “Electric Vehicles”

Integration of Computer Vision and IOT Into an Automatic Driving Assistance System for “Electric Vehicles”

Energy-Saving Driving Assistance System Integrated With Predictive Cruise Control for Electric Vehicles

Energy-Saving Driving Assistance System Integrated With Predictive Cruise Control for Electric Vehicles

Synchronous hybrid brain–computer interfaces for recognizing emergency braking intention

AbstractHybrid neurophysiological signals, such as the combination of electroencephalography (EEG) and electromyography (EMG), can be used to reduce road traffic accidents by obtaining the driver's intentions in advance and accordingly applying appropriate auxiliary controls. However, whether they can be used in combination and can achieve better results in situations of detecting emergency braking from normal driving and soft braking has not been explored. This study used one feature‐level (hybrid BCI‐FL) and three classifier‐level (hybrid BCIs‐CLs) hybrid strategies, the spectral band, and spectral point features to construct recognition models. Offline and pseudo‐online experiments were conducted. The recognition performance with the spectral point features showed a better result than that with spectral band features. In all experiments, the two proposed hybrid BCI strategies could achieve a detection accuracy close to or above 95%, while the detection advanced time is less than 300 ms. In particular, for the developed hybrid BCI recognition models, the hybrid BCI‐FL and hybrid BCI‐CL2 recognition models with spectral point features achieved 4.25% (p < 0.015) and 4.69% (p < 0.006) higher system accuracies, respectively, than that of the current better single EMG‐based recognition model. This research promotes the application of hybrid EEG and EMG signals in intelligent driving assistance systems.

Modelling and performance analysis of free body dynamics of electric vehicles

The modelling and performance analysis of free body dynamics in electric vehicles (EVs) plays a crucial role in understanding and optimizing the vehicle's behavior and performance. This research focuses on accurately modelling the forces and motions acting on the vehicle body, excluding the powertrain components. By considering factors such as vehicle weight, suspension characteristics, tire properties, and aerodynamic forces, a comprehensive mathematical model is developed. This model enables the simulation and analysis of the vehicle's behavior under various operating conditions. The performance analysis involves evaluating key metrics such as vehicle response, stability limits, and ride comfort. The proposed system is compared with the existing electric vehicle in the market. The findings from this research contribute to the design and development of EVs with improved handling, stability, and energy efficiency. Additionally, they inform the development of advanced driver-assistance systems (ADAS) and autonomous driving technologies. Overall, the modelling and performance analysis of free body dynamics in EVs supports the advancement of sustainable and efficient transportation systems.

Editorial

Dear Readers, In the fast-paced landscape, modern transportation is undergoing a profound transformation, driven by technological innovation, changing consumer preferences, and the need for sustainability. In the country, from urban areas to remote rural areas, the evolution of mobility technology is restructuring the way we navigate, unlocking new opportunities for efficiency, sustainability, and accessibility. As we stand at the threshold of a new era in mobility, it's essential to explore the innovative technologies driving this revolution and their potential to redefine the future of transportation. ARAI Journal of Mobility Technology takes pride in presenting its Jan–Mar 2024 issue that covers IC engine technology, Brake hose failure analysis, modeling on different technologies, battery and EV related subjects, and vehicle dynamics. The ARAI Journal journey is not just about getting the technologies which are about shaping the future of transportation and society as a whole. Electric vehicle technologies are poised to revolutionize the future of transportation, offering a cleaner, more sustainable alternative to traditional gasoline-powered cars. With growing concerns about climate change and environmental sustainability, the shift towards electric propulsion represents a significant step forward in reducing carbon emissions and mitigating the effects of climate change. With ongoing advancements in battery technology, charging infrastructure, and vehicle design, EVs are becoming increasingly practical and accessible for consumers around the world. By embracing innovation, fostering collaboration, and addressing the challenges ahead, we can accelerate the transition to a cleaner, greener transportation system that benefits both people and the planet. With ongoing advancements in battery technology, charging infrastructure, and vehicle design, EVs are becoming increasingly practical and accessible for consumers around the world. By embracing innovation, fostering collaboration, and addressing the challenges ahead, we can accelerate the transition to a cleaner, greener transportation system that benefits both people and beyond electrification, automotive technologies are also driving progress in safety and convenience. Advanced driver-assistance systems (ADAS) such as adaptive cruise control, lane-keeping assist, and automatic emergency braking are becoming standard features in modern vehicles, helping to prevent accidents and save lives. Moreover, the emergence of autonomous driving technology promises to revolutionize transportation by eliminating human error and making roads safer for everyone. ARAI Journal is a gateway for all researchers, industry professionals, students, faculty to discover the latest technologies and advancements driving the automotive industry forward on the planet. In the midst of life's challenges and uncertainties, may this journal be a source of solace and strength, reminding you of your resilience and inner wisdom. May it be a space where you can find clarity, purpose, and peace amidst the chaos of daily life. I want to extend my warmest wishes to each and every reader of this journal.

The evolution of embedded systems in automotive industry: A global review

The evolution of embedded systems in automotive industry: A global review

Driver Drowsiness Multi-Method Detection for Vehicles with Autonomous Driving Functions.

The article outlines various approaches to developing a fuzzy decision algorithm designed for monitoring and issuing warnings about driver drowsiness. This algorithm is based on analyzing EOG (electrooculography) signals and eye state images with the aim of preventing accidents. The drowsiness warning system comprises key components that learn about, analyze and make decisions regarding the driver's alertness status. The outcomes of this analysis can then trigger warnings if the driver is identified as being in a drowsy state. Driver drowsiness is characterized by a gradual decline in attention to the road and traffic, diminishing driving skills and an increase in reaction time, all contributing to a higher risk of accidents. In cases where the driver does not respond to the warnings, the ADAS (advanced driver assistance systems) system should intervene, assuming control of the vehicle's commands.

Fusion-GRU: A Deep Learning Model for Future Bounding Box Prediction of Traffic Agents in Risky Driving Videos

To ensure the safe and efficient navigation of autonomous vehicles and advanced driving assistance systems in complex traffic scenarios, predicting the future bounding boxes of surrounding traffic agents is crucial. However, simultaneously predicting the future location and scale of target traffic agents from the egocentric view poses challenges because the vehicles’ egomotion causes considerable field-of-view changes. Moreover, in anomalous or risky situations, tracking loss or abrupt motion changes limit the available observation time, requiring learning of cues within a short time window. Existing methods typically use a simple concatenation operation to combine different cues, overlooking their dynamics over time. To address this, this paper introduces the Fusion-Gated Recurrent Unit (Fusion-GRU) network, a novel encoder-decoder architecture for future bounding box localization. Unlike traditional gated recurrent units (GRUs), Fusion-GRU accounts for mutual and complex interactions among input features. Moreover, an intermediary estimator coupled with a self-attention aggregation layer is also introduced to learn sequential dependencies for long-range prediction. Finally, a GRU decoder is employed to predict the future bounding boxes. The proposed method is evaluated on two publicly available datasets, Risky Object Localization and Honda Egocentric Vehicle Intersection. The experimental results showcase the promising performance of the Fusion-GRU, demonstrating its effectiveness in predicting future bounding boxes of traffic agents.

The correlation between drivers’ road familiarity and glance behavior using real vehicle experimental data and mathematical models

Objective Road familiarity is an important factor affecting drivers’ visual features. Analyzing the quantitative correlation between drivers’ road familiarity and visual features in complex environment is of great help to improve driving safety. However, there are few relevant studies. This paper takes urban plane intersection as the environmental object to explore the correlation between drivers’ glance behavior and road familiarity, and conducts research on the quantitative evaluation model of road familiarity based on this correlation. Method First, a real vehicle experiment was carried out to record the eye movement data of 24 drivers with different road familiarity. The driver’s visual field plane was divided into 10 areas of interest (AOIs) based on the driver’s perspective. Three measures, including average glance duration, number of glances, and fixation transition probabilities between AOIs at urban plane intersections, were extracted. Finally, based on the experimental results, the driver road familiarity evaluation model was constructed using the factor analysis method. Results There are significant differences between unfamiliar and familiar drivers regarding the average glance duration toward the forward (FW) area, the left window (LW) area, the left rearview mirror (LVM) area and the left forward (LF) area, the number of glances toward the other (OT) area, and the fixation transition probabilities of LW→RF (right forward), LF→LF, LF→FW, FW→LW, FW→FW, FW→RVM (right rearview mirror). The comprehensive evaluation results show that the accuracy rate of the driver road familiarity evaluation model reached 83%. Conclusions This paper revealed that there is a strong correlation between drivers’ road familiarity and drivers’ glance behavior. Based on this correlation, we can include road familiarity as a part of drivers’ working status and establish a high accuracy evaluation model of driver road familiarity. The conclusion of this paper can provide some reference for the humanized design and improvement of advanced driving assistance system, which is of great significance for reducing the driving workload of drivers and improving the driving safety.

Driving emotions: using virtual reality to explore the effect of low and high arousal on driver’s attention

The role played by emotions and attention is crucial for the development of advanced driver assistance systems that improve safety by flexibly adapting to the current state of the driver. In the present study, we used immersive virtual reality as a testing tool to investigate how different emotional states affect drivers’ attention in a divided attention task. Two different emotional states, diversified by valence and arousal, were induced before performing a divided attention task in a driving simulation. The experimental task developed for this study allowed us to explore if and how two different emotional states can affect the way drivers divide their attention between a central driving-related task and a peripheral visual task. Our results showed that scared drivers presented lower reaction times at the central task compared to relaxed drivers. On the contrary, the emotional state did not affect the performance at the peripheral task, which revealed instead a significant effect of the eccentricity at which the visual stimuli were presented, influencing both the accuracy of targets’ perception and participants’ reaction times.

An advanced multimodal driver-assistance prototype for emergency-vehicle detection

In the automotive industry, intelligent monitoring systems for advanced human-vehicle interaction aimed at enhancing the safety of drivers and passengers represent a rapidly growing area of research. Safe driving behavior relies on the driver’s awareness of the road context, enabling them to make appropriate decisions and act consistently in anomalous circumstances. A potentially dangerous situation can arise when an emergency vehicle rapidly approaches with sirens blaring. In such cases, it is crucial for the driver to perform the correct maneuvers to prioritize the emergency vehicle. For this purpose, an Advanced Driver Assistance System (ADAS) can provide timely alerts to the driver about an approaching emergency vehicle. In this work, we present a driver-assistance prototype that leverages multimodal information from an integrated audio and video monitoring system. In the initial stage, sound analysis technologies based on computational audio processing are employed to recognize the proximity of an emergency vehicle based on the sound of its siren. When such an event occurs, an in-vehicle monitoring system is activated, analyzing the driver’s facial patterns using deep-learning-based algorithms to assess their awareness. This work illustrates the design of such a prototype, presenting the hardware technologies, the software architecture, and the deep-learning algorithms for audio and video data analysis that make the driver-assistance prototype operational in a commercial car. At this initial experimental stage, the algorithms for analyzing the audio and video data have yielded promising results. The area under the precision-recall curve for siren identification stands at 0.92, while the accuracy in evaluating driver gaze orientation reaches 0.97. In conclusion, engaging in research within this field has the potential to significantly improve road safety by increasing driver awareness and facilitating timely and well-informed reactions to crucial situations. This could substantially reduce risks and ultimately protect lives on the road.

A review of driver gaze estimation and application in gaze behavior understanding

Driver gaze plays a key role in different gaze-based applications, such as driver attentiveness detection, visual distraction detection, gaze behavior understanding, and building driver assistance system. The main objective of this study is to perform a comprehensive summary of driver gaze fundamentals, methods to estimate driver gaze using machine learning (ML) based technique, and its applications in real world driving scenarios. We first discuss the fundamentals related to driver gaze, involving head-mounted and remote setup based gaze estimation and the terminologies used for driver gaze behavior understanding. Next, we list out the existing benchmark driver gaze datasets, highlighting the collection methodology, and the equipment used for such data collection. This is followed by a discussion of the algorithms used for driver gaze estimation, which primarily involves traditional machine learning and deep learning (DL) based techniques. The estimated driver gaze is then used for understanding gaze behavior while maneuvering through intersections, on-ramps, off-ramps, lane changing, determining the effect of roadside advertising structures and also for developing driver gaze based applications such as maneuver prediction, driver inattention, and distraction detection systems, etc. Finally, we discuss the limitations in the existing literature, challenges, and future scope in driver gaze estimation and gaze-based applications.

A comprehensive high‐level automated driving assistance system with integrated multi‐functionality

AbstractAdvanced Driver Assistance Systems (ADAS) have gained substantial attention in recent years. However, the integration mechanism of multiple functions within ADAS remains unexplored, and the full potential of its functionality remains underutilised. This paper presents a novel multi‐functional integrated High‐level Automated Driving Assistance System that combines the Cruise Control (CC), Adaptive Cruise Control (ACC), Automated Emergency Brake (AEB), and Automated Lane Change (ALC) functions. The presented system utilises a hierarchical framework. The extension multi‐mode switch strategy is established as the superior module and the Event‐Triggered Model Predictive Controller (ETMPC) is designed as the inferior controller. The CC, ACC, and ALC functions are effectively utilised to enhance traffic efficiency, while the AEB function ensures driving safety. To address the time constraints of conventional Model Predictive Control, an event‐trigger mechanism is proposed to reduce computational load. Simulations are conducted using the CarSim and Matlab platforms. The study results demonstrate significant improvements in both safety and traffic efficiency compared to conventional ADAS strategies. Furthermore, the proposed ETMPC method significantly reduces the frequency of solving Optimisation Problems and decreases online computation costs.

AI monitoring and warning system for low visibility of freeways using variable weight combination model

AbstractIn intelligent vehicles, road environment perception technology is a key component of autonomous driving assistance systems. This component is the foundation for vehicle decision‐making and control, and is a guarantee of safety during the driving of the vehicle. The existing environment perception technology mainly targets well‐lit environments and requires visible light imaging equipment. Therefore, in low visibility environments, this technology cannot make good judgments about the external environment. Many existing perception systems mainly rely on sensors. Under low visibility conditions, these sensors weaken their effectiveness due to signal transmission, reflection, or absorption, resulting in incomplete or distorted data collection. Reduced visibility often affects the sensing range of various sensors, hindering the system's ability to detect and recognize distant objects, thereby limiting the necessary advance warning and response time for safe navigation. In response to this issue, this study proposed a combined method of infrared imaging and polarized imaging to collect feature data on road conditions in low visibility environments. Then, the obtained images were denoised and enhanced. The processed images were input into the system for recognition, and the images were analyzed and recognized using a low visibility road situation semantic segmentation algorithm based on deep learning. The research outcomes denoted that the pixel accuracy, average pixel accuracy, and average intersection ratio of the variable weight combination model in polarized degree images were 91.2%, 89.1%, and 71.6%, respectively. Those in infrared images were 83.6%, 90.6%, and 62.1%, respectively. The various indicators of the variable weight combination model were higher than those of the U‐shaped neural network model, indicating its performance is relatively excellent. The research results indicated that infrared imaging helps to acquire information at night or in low light conditions, while polarized imaging can provide better adaptation to cluttered light and reflections, enabling the system to provide more robust environmental sensing in complex weather conditions. It fills a critical gap in perception for autonomous driving systems in adverse weather conditions.

Modeling the influence of connected vehicles on driving behaviors and safety outcomes in highway crash scenarios across varied weather conditions: A multigroup structural equation modeling analysis using a driving simulator experiment

Equipped with advanced sensors and capable of relaying safety messages to drivers, connected vehicles (CVs) hold the potential to reduce crashes. The goal of this study is to assess the impacts of CV technologies on driving behaviors and safety outcomes in highway crash scenarios under diverse weather conditions, including clear and foggy weather. A driving simulator experiment was conducted and the multigroup structural equation modeling (SEM) was employed to explore the complex interrelationships between the propensity of traffic conflicts, utilization of CV alerts, weather, psychological factors, driving behaviors, and other relevant variables for two different crash locations, namely a straight section and a horizontal curve. Two latent psychological factors including aggressiveness and unawareness were constructed from driving behavior as vehicles passed by crash scenes such as brake, throttle, steering angle, lane offset, and yaw. The SEM can measure latent psychological factors and model interrelationships concurrently through a single statistical estimation procedure. Results of the multigroup SEM showed that CV alerts could significantly reduce the unawareness on a horizontal curve and thus lower the propensity of traffic conflicts. Additionally, the overall effect of foggy weather on conflicts was found to be positive on a horizontal curve, despite the potential benefit of improving situational awareness. In contrast, the single group SEM failed to reveal any significant interrelationships in its structural model by pooling data from both crash locations. The obtained insights can guide the development of driving assistance systems, highlighting the necessity of customization considering weather conditions and location-specific factors.

Behavioral Patterns of Drivers under Signalized and Unsignalized Urban Intersections

Under the general trend of mixed traffic flow, an in-depth understanding of the driving behaviors of traditional vehicles is of great significance for the design of autonomous vehicles and the improvement in the safety and acceptance of autonomous vehicles. This study first obtained microdata on the behaviors of drivers through driving simulation experiments and conducted research in stages. Then, generalized linear mixed-effects models were constructed to study the main effects and interaction effects of driver attributes and traffic conditions on driving behaviors. The data analysis shows that the overall speed of drivers passing through intersections follows a “deceleration acceleration” mode, but the fluctuations are more pronounced at signalized intersections, and the signal control significantly changes the position of the lowest speed when turning left. According to the different signal control and driving tasks, there are significant differences in a driver’s acceleration patterns between the entry and exit stages. A driver’s heart rate fluctuates greatly during the exit phase, especially during straight tasks. Compared with other indicators, the change in the gaze duration is not significant. In addition, interaction effects were observed between driver attributes and traffic conditions, with participants exhibiting different behavioral patterns based on their different attributes. The research results can provide a basis for the design of driving assistance systems and further improve the interactions between autonomous vehicles and traditional vehicles at intersections.

VELIE: A Vehicle-Based Efficient Low-Light Image Enhancement Method for Intelligent Vehicles.

In Advanced Driving Assistance Systems (ADAS), Automated Driving Systems (ADS), and Driver Assistance Systems (DAS), RGB camera sensors are extensively utilized for object detection, semantic segmentation, and object tracking. Despite their popularity due to low costs, RGB cameras exhibit weak robustness in complex environments, particularly underperforming in low-light conditions, which raises a significant concern. To address these challenges, multi-sensor fusion systems or specialized low-light cameras have been proposed, but their high costs render them unsuitable for widespread deployment. On the other hand, improvements in post-processing algorithms offer a more economical and effective solution. However, current research in low-light image enhancement still shows substantial gaps in detail enhancement on nighttime driving datasets and is characterized by high deployment costs, failing to achieve real-time inference and edge deployment. Therefore, this paper leverages the Swin Vision Transformer combined with a gamma transformation integrated U-Net for the decoupled enhancement of initial low-light inputs, proposing a deep learning enhancement network named Vehicle-based Efficient Low-light Image Enhancement (VELIE). VELIE achieves state-of-the-art performance on various driving datasets with a processing time of only 0.19 s, significantly enhancing high-dimensional environmental perception tasks in low-light conditions.