Vehicles

Driving at night is riskier in terms of crash involvement than it is during the day. Fortunately, it is clearly established that illumination on roadways can reduce the number and severity of nighttime crashes. However, state and municipal departments of transportation (DOTs) lack the available illumination data. Therefore, the objective of this research is threefold, as follows: (i) to develop machine learning models that use readily available roadway characteristic data to predict the severity of nighttime crashes; (ii) determine the effect that illumination has on crash severity; and (iii) develop a tool to assist DOT decision makers in collecting illumination data. To accomplish this objective, we have extracted data from the Texas Department of Transportation (TxDOT) Crash Record Information System (CRIS) database, which was then further split into a training and a test dataset. Then, seven machine learning techniques, namely binary logistic regression, k-nearest neighbors, naïve Bayes, random forest, artificial neural network, Extreme Gradient Boosting (XGBoost), and a Long Short-Term Memory (LSTM) model, were all applied to the unseen test data. The random forest model produced the most promising results by predicting severe crashes with 97.6% accuracy. In addition, we conducted a pilot study to test the collection of illumination data using a light meter. In the future, we aim to complete the development of a smartphone application, which can be used in conjunction with the random forest model presented in this paper, to collect crowdsourced illumination data and predict nighttime crash hotspots. This may assist DOT decision makers to prioritize funding for illumination at the hot spots. Full article

The electrified environments encountered in electric vehicles (EVs) in terms of parasitic currents present significant challenges for the performance of EV bearings and their lubricants. This study investigates the effectiveness of various concentrations (0.1 wt.%, 0.2 wt.%, 0.3 wt.%, and 0.4 wt.%) of multi-walled carbon nanotubes (MWCNT) and alumina (Al₂O₃) as two different nanoparticles incorporated into lithium grease, specifically focusing on their ability to mitigate the bearing surface damage caused by varying magnitudes of bearing DC discharges. A specialized test rig was developed to evaluate the electrical discharge characteristics, vibration response, and extent of surface wear on bearings lubricated with both lithium grease without additives and when infused with each nano-additive. Microscopic examination was employed to qualitatively and quantitatively evaluate the surface degradation of each test bearing. The results of this study demonstrate that the addition of nano-additives into the lubricating grease of bearings subjected to electrical loads resulted in a reduction in electric discharge voltage thresholds and levels. This reflected on the mitigation of surface damage in terms of surface roughness and vibration amplitudes by up to 70.67% and 65.19% in the case of MWCNTs. In contrast, alumina nanoparticles yielded a reduction in vibration amplitude and surface wear by 44.89% and 37.5%, respectively. Full article

When a railway train runs along a curved track with braking, the dynamic behaviors of the vehicle are extremely complex and difficult to accurately reveal due to the coupling effects between the wheel–rail interactions and the disc–pad frictions. Therefore, a rigid–flexible coupled trailer car dynamics model of a railway train is established. In this model, the brake systems and vehicle system are dynamically coupled via the frictions within the braking interface, wheel–rail relationships and suspension systems. Furthermore, the effectiveness of the established model is validated by a comparison with the field test data. Based on this, the dynamic response characteristics of vehicle under curve and straight braking conditions are analyzed and compared, and the influence of the curve geometric parameters on vehicle vibration and operation safety is explored. The results show that braking on a curve track directly affects the vibration characteristics of the vehicle and reduces its operation safety. When the vehicle is braking on a curve track, the lateral vibration of the bogie frame significantly increases compared to the vehicle braking on a straight track, and the vibration intensifies as the curve radius decreases. When the curved track maintains equilibrium superelevation, the differences in primary suspension force, wheel–rail vertical force, and wheel axle lateral force between the inner and outer sides of the first and second wheelsets are relatively minor under both straight and curved braking conditions. Additionally, under these circumstances, the derailment coefficient is minimized. However, when the curve radius is 7000 m, with a superelevation of 40 mm, the maximum dynamic wheel load reduction rate of the inner wheel of the second wheelset is 0.54, which reaches 90% of the allowable limit value of 0.6 for the safety index, and impacts the vehicle running safety. Therefore, it is necessary to focus on the operation safety of railway trains when braking on curved tracks. Full article

Given the considerable length and weight of freight trains, their operation can be quite challenging. Improper operation may lead to train decoupling and derailment. Driver Advisory Systems (DASs) are used in some countries to assist train drivers by providing the speed curves, which are desired to be easy to track. Multi-mass train model is a good choice to depict the in-train forces in train speed curve generating, but its application is often hindered by the computation time. A single mass train model is considered as another choice to simplify the computation. To exploit the advantages of the multi-mass and single-mass models, this paper proposes a Two-step Optimization Method to generate the optimal speed curves for the freight trains. In the first step, the Rolling Optimization Algorithm (ROA) is proposed to optimize the speed curve on the basis of the single-mass model, taking the train energy consumption and punctuality as the optimization objectives. In order to assist the driver in operating the train smoothly, the speed curve generated by the ROA was tested on DAS, but it could not be followed accurately in the actual operation. To solve this problem, a Model Predictive Control (MPC) algorithm based on a multi-mass model is adopted as the second optimization step, which takes the output speed curve of the ROA as the reference speed curve. The MPC algorithm will generate a new speed curve, taking in-train forces, energy consumption and punctuality as the optimization indices. Simulations are carried out using the data from the Dalailong railway in China to evaluate the proposed method. The simulation results show that the speed curves generated by the Two-step Optimization Method are smoother than that of the ROA, and the throttle sequences are more conducive for the driver to follow in practical operation. The simulation results show that the energy consumption is reduced by 17.1% compared to that of the ROA simulation. The speed curve also can be integrated into the onboard DAS or the Automatic Train Operation (ATO) system, aiming to obtain a smooth and energy-efficient train operation. Full article

In the global context, transportation contributes 26% of the total CO₂ emissions, with land transport responsible for 92% of the emissions within the sector. Given this significant contribution to climate change, it is crucial to quantify vehicular impacts to implement effective mitigation strategies. This study introduces an innovative method for predicting fuel consumption and emissions of carbon monoxide, hydrocarbons, and nitrogen oxides in vehicles, based on instantaneous vehicle-specific power (VSP) and mean accumulated power. VSP is a parameter that measures a vehicle’s power in relation to its mass, providing an indicator of the efficiency with which the vehicle converts fuel into motion. This indicator is particularly useful for assessing how vehicles utilize their energy under different driving conditions and how this affects their fuel consumption and emissions. Using data collected from 10 vehicles over 2000 h and covering altitudes from 0 to 4000 m above sea level in Ecuador, the method not only improved the accuracy of consumption predictions, reducing the margin of error by up to 10% at high altitudes, but also provided a detailed understanding of how altitude affects both consumption and emissions. The precision of the new method was notable, with a standard deviation of only 0.25 L per 100 km, allowing for reliable estimates under various operational conditions. Interestingly, the study revealed an average increase in fuel consumption of 0.43 L per 1000 m of altitude gain, while CO₂ emissions showed a significant reduction from 260.93 g/km to 215.90 g/km when ascending from 500 m to 4000 m. These findings underscore the relevance of considering altitude in route planning, especially in mountainous terrains, to optimize performance and environmental sustainability. However, the study also indicated an increase in CO and NOx emissions with altitude, a challenge that highlights the need for integrated strategies addressing both fuel consumption and air quality. Collectively, the results emphasized the complex interplay between altitude, energy efficiency, and vehicular emissions, underscoring the importance of a holistic approach to transportation management, to minimize adverse environmental impacts and promote sustainability. Full article

Induction motor (IM) drives are considered one of the important technologies in modern industry. Several industrial applications, such as material handling and food and beverage applications, are driven and operated by modern AC drives. Moreover, modern electric transportation systems such as EVs and e-trucks are based on AC drives. Recently, high-efficiency IM drive systems have been studied as a major opportunity to reduce energy and fuel consumption. This article addresses the recent trends and advancement in high-efficiency IM drives during a particular period (2017–2024), including the development of high-efficiency motors, the utilization of efficient wide bandgap (WBG) semiconductor devices for inverter topology, and commonly used control strategies to achieve high-performance drives. Moreover, the article addresses several manufacturers of industrial IM drives and the corresponding adopted control techniques in their products. A comparison of these control techniques, including their pros and cons, has been conducted as well. Full article

Sliding Mode Control (SMC) has gained significant attention due to its simplicity, robustness, and rapid response in ensuring system stability, particularly with the Lyapunov approach. Despite its advantages, SMC faces challenges such as chattering near equilibrium, sensitivity to parameter variations, and delayed convergence. To address these issues, advanced techniques like Terminal Sliding Mode Control (TSMC) and Integral Terminal Sliding Mode Control (ITSMC) have been proposed. TSMC ensures finite-time convergence while mitigating chattering, while ITSMC further handles singularities and disturbances. Additionally, Adaptive Switching Control (ASC) based on Particle Swarm Optimization (PSO) is applied to achieve faster convergence, suppress chattering, and enhance system robustness. The adaptive control law, utilizing a Lyapunov-based approach, is employed to estimate and compensate for external disturbances, further improving system performance under uncertainties. Gain tuning, essential for optimizing system performance and reducing tracking errors, is achieved using the efficient Teaching–Learning-Based Optimization (TLBO) algorithm. This study applies TSMC, ITSMC, and ASC-based PSO to an Anti-Lock Braking System (ABS), aiming to enhance robustness, stability, and finite-time convergence while reducing chattering. Stability is analyzed through the Lyapunov theory, ensuring rigorous validation. MATLAB simulations demonstrate the effectiveness of the proposed methods in improving ABS performance, offering a valuable contribution to robust control techniques for systems operating under dynamic and uncertain conditions. Full article

This manuscript provides actionable recommendations to enhance user satisfaction and address existing barriers regarding pooled rideshare (PR) in the United States. Despite PR’s intended benefits, such as reduced traffic congestion and cost savings, its adoption remains limited. To identify these actionable items, a U.S. nationwide survey with 5385 participants explored transportation preferences, barriers, and motivators for PR use in the summer of 2021. First, two factor analyses were conducted. The first factor analysis identified the five factors associated with one’s willingness to consider PR (time/cost, traffic/environment, safety, privacy, and service experience). The second factor analysis revealed the four factors related to ways to optimize one’s PR experience (comfort/ease of use, convenience, vehicle technology/accessibility, and passenger safety). Privacy concerns, for instance, were found to reduce the likelihood of PR adoption by 77%, and convenience had the potential to increase it by 156%. A structural equation model evaluated the relationships among these nine key factors influencing PR usage to develop the Pooled Rideshare Acceptance Model (PRAM). The privacy, safety, trust service, and convenience factors each had a significant large effect (Cohen’s f² > 0.35) on the model. PRAM was extended using multigroup analyses to reveal the nuanced impact of 16 demographics, including gender, generation, rideshare experience, etc., highlighting the need for tailored strategies to improve PR acceptance through the Pooled Rideshare Acceptance Model Multigroup Analyses (PRAMMAs). Multiple workshops were held with diverse audiences to translate the team’s findings to date into 84 actionable recommendations, categorized across topical areas like safety, routing, driver and passenger selection, user education, etc. These findings are a foundation for a future study to determine which items resonate with different user groups. In the meantime, the actional items serve as a user-driven resource for policymakers, transportation network companies, and researchers, offering a roadmap to potential improvements to PR services to address existing concerns with the goal of increasing the usage of PR. Full article

With the rapid development of the Internet of Vehicles (IoV) and autonomous driving technologies, robust and accurate visual pose perception has become critical for enabling smart connected vehicles. Traditional deep learning-based localization methods face persistent challenges in real-world vehicular environments, including occlusion, lighting variations, and the prohibitive cost of collecting diverse real-world datasets. To address these limitations, this study introduces a novel approach by combining Vision-LSTM (ViL) with synthetic image data generated from high-fidelity 3D models. Unlike traditional methods reliant on costly and labor-intensive real-world data, synthetic datasets enable controlled, scalable, and efficient training under diverse environmental conditions. Vision-LSTM enhances feature extraction and classification performance through its matrix-based mLSTM modules and advanced feature aggregation strategy, effectively capturing both global and local information. Experimental evaluations in independent target scenes with distinct features and structured indoor environments demonstrate significant performance gains, achieving matching accuracies of 91.25% and 95.87%, respectively, and outperforming state-of-the-art models. These findings underscore the innovative advantages of integrating Vision-LSTM with synthetic data, highlighting its potential to overcome real-world limitations, reduce costs, and enhance accuracy and reliability for connected vehicle applications such as autonomous navigation and environmental perception. Full article

This study introduces a novel approach for traffic control systems by using Large Language Models (LLMs) as traffic controllers. The study utilizes their logical reasoning, scene understanding, and decision-making capabilities to optimize throughput and provide feedback based on traffic conditions in real time. LLMs centralize traditionally disconnected traffic control processes and can integrate traffic data from diverse sources to provide context-aware decisions. LLMs can also deliver tailored outputs using various means such as wireless signals and visuals to drivers, infrastructures, and autonomous vehicles. To evaluate LLMs’ ability as traffic controllers, this study proposed a four-stage methodology. The methodology includes data creation and environment initialization, prompt engineering, conflict identification, and fine-tuning. We simulated multi-lane four-leg intersection scenarios and generated detailed datasets to enable conflict detection using LLMs and Python simulation as a ground truth. We used chain-of-thought prompts to lead LLMs in understanding the context, detecting conflicts, resolving them using traffic rules, and delivering context-sensitive traffic management solutions. We evaluated the performance of GPT-4o-mini, Gemini, and Llama as traffic controllers. Results showed that the fine-tuned GPT-mini achieved 83% accuracy and an F1-score of 0.84. The GPT-4o-mini model exhibited a promising performance in generating actionable traffic management insights, with high ROUGE-L scores across conflict identification of 0.95, decision making of 0.91, priority assignment of 0.94, and waiting time optimization of 0.92. This methodology confirmed LLMs’ benefits as a traffic controller in real-world applications. We demonstrated that LLMs can offer precise recommendations to drivers in real time including yielding, slowing, or stopping based on vehicle dynamics. This study demonstrates LLMs’ transformative potential for traffic control, enhancing efficiency and safety at intersections. Full article

Given the complexity of the crashes and the increasing interest in public policies related to the reduction in both accidents and fatalities from road crashes, the proposed review of the specialty literature may serve as a starting point for individuals interested in developing studies related to road vehicle accidents, reconstruction methodologies, assessment of vehicles crashworthiness, as well as evaluation of occupants’ behavior in different collision scenarios. Therefore, the present paper aims to offer a comprehensive overview of the specialty literature approaches in terms of road vehicle accidents through an analysis of the reconstruction methods used in the cases of vehicle-to-vehicle or vehicle-to-object crashes, as well as ways in which the crashworthiness of road vehicles is assessed by specialized organizations or individual experts. The addressed topics were summarized from a range of European and global strategies in the field of transportation, reports, testing protocols, as well as scientific research papers published in international databases. The main purpose of the present paper is to serve as a foundational resource for researchers and practitioners seeking to contextualize their work within a global framework. Full article

The paper discusses the potential for autonomous vehicles to improve traffic flow on roundabouts, suggesting that their ability to slow down strategically can enhance traffic and reduce pollution on both main and yielding roads. A traffic simulator for a roundabout was developed for a busy intersection of a new city neighborhood. We consider that some of the cars are self-driving, and they are fully aware of the traffic scenario. By optimizing their speed and timing their speed reduction, these vehicles can help maintain a balance between the number and time of crossing vehicles on both the main and yielding roads. This study evaluates the effectiveness of the intervention, demonstrating that autonomous vehicles can significantly improve roundabout efficiency, reducing congestion and pollution. The application of genetic algorithms is highlighted as an effective optimization method to find the right autonomous vehicle’s timing and speed reduction ratio combination on the main road to enhance traffic efficiency. Full article

The increasing integration of autonomous driving systems into modern vehicles heightens the significance of Image Quality Assessment (IQA), as it pertains directly to vehicular safety. In this context, the development of metrics that can emulate the Human Visual System (HVS) in assessing image quality assumes critical importance. Given that blur is often the primary aberration in images captured by aging or deteriorating camera sensors, this study introduces a No-Reference (NR) IQA model termed BREMOLA (Blind/Referenceless Model via Moving Spectrum and Laplacian Filter). This model is designed to sensitively respond to varying degrees of blur in images. BREMOLA employs the Fourier transform to quantify the decline in image sharpness associated with increased blur. Subsequently, deviations in the Fourier spectrum arising from factors such as nighttime lighting or the presence of various objects are normalized using the Laplacian filter. Experimental application of the BREMOLA model demonstrates its capability to differentiate between images processed with a 3 × 3 average filter and their unprocessed counterparts. Additionally, the model effectively mitigates the variance introduced in the Fourier spectrum due to variables like nighttime conditions, object count, and environmental factors. Thus, BREMOLA presents a robust approach to IQA in the specific context of autonomous driving systems. Full article

The electric traction drive is increasingly being applied as a device providing traction force on driving wheels. This is due to its reliable torque transmission to the driving wheels, step-less regulation of the traction force on the driving wheels depending on the driving conditions, and increased design capabilities. In terms of power, the electric traction drive has maximum torque at low speeds, which internal combustion engines lack. This property of the electric drive is not applied in urban vehicles, as not all passengers are comfortable with intensive acceleration. In modern vehicles with an electric traction drive, the maximum acceleration can be limited by software, which is the focus of this study. This paper aims to develop an algorithm capable of recognizing when the permissible longitudinal acceleration exceeds the limit and generating an action to maintain the acceptable acceleration level. The electric traction drive of a large-class electric bus was used as a control object. An algorithm and a control law are hereby developed, which reduce longitudinal acceleration using PI control. Both simulation modeling and full-scale tests on the electric bus were carried out to evaluate the performance and efficiency of the algorithm. In this paper, the authors also introduce the cumulative velocity concept and prove the operability and efficiency of the developed method. Full article

The energy storage system (ESS) plays a crucial role in electric vehicles (EVs), impacting their performance and efficiency. While batteries are the standard choice for energy storage, they come with drawbacks like low power density and limited life cycles, which can hinder pure battery electric vehicles (PBEVs). To address these issues, a hybrid energy storage system (HESS) that combines a battery with a supercapacitor provides a more effective solution. The battery delivers consistent power, while the supercapacitor manages peak power demands and regenerative braking energy. This study proposes a new energy management strategy for the HESS, an advanced adaptive rule-based algorithm. The results of the standard rule-based and adaptive rule-based algorithms are used to verify the proposed control algorithm. The system was modeled in MATLAB/Simulink and evaluated across three driving cycles—UDDS, NYCC, and Japan1015—while varying states of charge for the supercapacitors. The findings indicate that the HESS significantly alleviates battery stress compared to a pure battery system, enhancing both efficiency and lifespan. Among the algorithms tested, the advanced adaptive rule-based algorithm yielded the best results, increasing the number of viable drive cycles. Full article

Traffic accident-related injuries and fatalities are a serious global public health and social development challenge. The accurate prediction of crash severity improves road safety and reduces casualties, as well as serving road managers and policy makers. Prediction models need to learn and analyze the various characteristic factors of traffic accidents and capture from them the inherent complex relationship between accident characteristics and the severity of traffic accidents. However, most accident prediction studies lack analytical predictions of injury severity, and predictive models rely on the content and quality of accident datasets. To increase the robustness and accuracy of prediction models, this paper leverages a Transformer-based architecture for the severity prediction of traffic collisions from human injury severity. This framework learns both text and sequence data from accident datasets. After comparative analysis, the framework can achieve the prediction of human injury severity under different data categories and show good prediction performance at low injury severity levels using only textual data or sequence data. Full article

This paper addresses the challenge of improving decision-making capabilities and safety in autonomous vehicles (AVs) using Agent-Based Modelling (ABM). The study evaluates ABM’s effect on Advanced Driver Assistance Systems (ADASs) in challenging driving situations, like lane merging, by incorporating it into a simulation framework designed for autonomous vehicles. Identifying emergent behaviours that enhance safety and efficiency, verifying the efficacy of ABM in AV decision-making, and investigating the function of hardware acceleration to enable practical application in ADASs are some of the major achievements. According to the simulation results, ABM can greatly improve AV performance, providing a practical and scalable means of enhancing safety in future transportation systems. Full article

Driving fatigue is the cause of many traffic accidents and poses a serious threat to road safety. To address this issue, this paper aims to develop a system for the early detection of driver fatigue. The system leverages heart rate variability (HRV) features and embedded machine learning to estimate the driver’s fatigue level. The driver’s HRV is derived from electrocardiogram (ECG) signals captured by a wearable device for analysis. Time- and frequency-domain HRV features are then extracted and used as the input for a machine learning classifier. A dataset of HRV features is collected from a driving simulation experiment involving 18 participants. Four machine learning classifiers are evaluated, and a backpropagation neural network (BPNN) is selected for its superior performance, achieving up to 94.35% accuracy. The optimized classifier is successfully deployed on an embedded system, providing a cost-effective and portable solution for the early detection of driver fatigue. The results demonstrate the feasibility of using HRV-based machine learning models for the early detection of driver fatigue, contributing to enhanced road safety and a reduced accident risk. Full article

Lane detection is a pivotal technology of the intelligent driving system. By identifying the position and shape of the lane, the vehicle can stay in the correct lane and avoid accidents. Image-based deep learning is currently the most advanced method for lane detection. Models using this method already have a very good recognition ability in general daytime scenes, and can almost achieve real-time detection. However, these models often fail to accurately identify lanes in challenging scenarios such as night, dazzle, or shadows. Furthermore, the lack of diversity in the training data restricts the capacity of the models to handle different environments. This paper proposes a novel method to train CycleGAN with existing daytime and nighttime datasets. This method can extract features of different styles and multi-scales, thereby increasing the richness of model input. We use CycleGAN as a domain adaptation model combined with an image segmentation model to boost the model’s performance in different styles of scenes. The proposed consistent loss function is employed to mitigate performance disparities of the model in different scenarios. Experimental results indicate that our method enhances the detection performance of original lane detection models in challenging scenarios. This research helps improve the dependability and robustness of intelligent driving systems, ultimately making roads safer and enhancing the driving experience. Full article