Vehicle Speed Research Articles

From Stationary to Nonstationary UAVs: Deep-Learning-Based Method for Vehicle Speed Estimation

The development of smart cities relies on the implementation of cutting-edge technologies. Unmanned aerial vehicles (UAVs) and deep learning (DL) models are examples of such disruptive technologies with diverse industrial applications that are gaining traction. When it comes to road traffic monitoring systems (RTMs), the combination of UAVs and vision-based methods has shown great potential. Currently, most solutions focus on analyzing traffic footage captured by hovering UAVs due to the inherent georeferencing challenges in video footage from nonstationary drones. We propose an innovative method capable of estimating traffic speed using footage from both stationary and nonstationary UAVs. The process involves matching each pixel of the input frame with a georeferenced orthomosaic using a feature-matching algorithm. Subsequently, a tracking-enabled YOLOv8 object detection model is applied to the frame to detect vehicles and their trajectories. The geographic positions of these moving vehicles over time are logged in JSON format. The accuracy of this method was validated with reference measurements recorded from a laser speed gun. The results indicate that the proposed method can estimate vehicle speeds with an absolute error as low as 0.53 km/h. The study also discusses the associated problems and constraints with nonstationary drone footage as input and proposes strategies for minimizing noise and inaccuracies. Despite these challenges, the proposed framework demonstrates considerable potential and signifies another step towards automated road traffic monitoring systems. This system enables transportation modelers to realistically capture traffic behavior over a wider area, unlike existing roadside camera systems prone to blind spots and limited spatial coverage.

System to Assist the Driver During a Single Lane Change Maneuver, in the Conditions of Danger Arising from a Change in the Condition of the Road Surface

The article describes the application of a vehicle simulation model to assess the driving hazards that arise due to changes in the tyre–road adhesion during a lane change. An experimentally verified vehicle dynamics model was used. The typical single lane change maneuver, performed on dry concrete, wet concrete, and ice-covered roads, was simulated for three vehicle speeds. The disturbance was introduced as a changed tyre–road adhesion on the target lane. Typical sinusoidal (one-period) input was applied to the steering wheel. Adhesion change may lead to an accident when considering the driver’s reaction time. An original control algorithm has been built. It is the driver assistance system with a double PID controller of the steering wheel angle. The system parameters were determined based on the principles recommended in the automatic control engineering monographs. The controller fulfils its tasks for motion at very high speeds on a homogeneous road surface and for the case where the tyre–road adhesion is changed during a maneuver. Thanks to this, the threat can be avoided.

Degradation-considering adaptive equivalent consumption minimization strategy for fuel cell electric vehicles

In order to improve the economy and durability of fuel cell vehicle, a degradation-considering adaptive equivalent consumption minimization strategy (D-ECMS) was proposed for fuel cell/lithium battery hybrid power system. Aiming at enhancing the adaptability of the equivalent consumption minimum strategy, four different vehicle speed conditions were considered including low speed, medium speed, high speed, and ultra-high speed. Particle swarm optimization (PSO) algorithm was used to optimize the penalty factor under different vehicle speed conditions, and BP neural network was used to identify real-time conditions to form an adaptive equivalent consumption minimum strategy (A-ECMS). The voltage decay model of the fuel cell was established according to the voltage decay rate of the fuel cell under the three adverse conditions of idle speed, high power and dynamic load cycle. The voltage decay of the fuel cell was transformed to hydrogen consumption in the cost function of the A-ECMS strategy to obtain the D-ECMS considering durability degradation. The simulation and the hardware-in-the-loop (HIL) test based on the dSPACE real-time platform was carried out. The results show that: compared with A-ECMS, the hydrogen consumption of the D-ECMS increased by 5.2%, 9.7%, 10.2%, and 0.2% respectively under WLTC, NEDC, UDDS, and HWFET cycles. However, D-ECMS can reduce the voltage decay by 38.3%, 59.7%, 35.2%, and 26% under the above four test conditions, and their total comprehensive cost loss is reduced by 8.7%, 6.7%, 9.6%, and 6.3% respectively. As result, the D-ECMS can reduces the voltage attenuation of fuel cells and improve the durability of fuel cells significantly with a small hydrogen consumption increase,which can reduce the operating cost of fuel cells finally. Furthermore, the average error between the results of HIL test and offline simulation results is less than 5%, which validates the adaptability and real-time application feasibility of D-ECMS.

Exploring Factors Influencing Speeding on Rural Roads: A Multivariable Approach

Speeding is one of the main contributing factors to road crashes and their severity; therefore, this study aims to investigate the complex dynamics of speeding and uses a multivariable analysis framework to explore the diverse factors contributing to exceeding vehicle speeds on rural roads. The analysis encompasses diverse measured variables from Croatia’s secondary road network, including time of day and supplementary data such as average summer daily traffic, roadside characteristics, and settlement location. Measuring locations had varying speed limits ranging from 50 km/h to 90 km/h, with traffic volumes from very low to very high. In this study, modeling of influencing factors on speeding was carried out using conventional and more advanced methods with speeding as a binary dependent variable. Although all models showed accuracy above 74%, their sensitivity (predicting positive cases) was greater than specificity (predicting negative cases). The most significant factors across the models included the speed limit, distance to the nearest intersection, roadway width, and traffic load. The findings highlight the relationship between the variables and speeding cases, providing valuable insights for policymakers and law enforcement in developing measures to improve road safety by determining locations where speeding is expected and planning further measures to reduce the frequency of speeding vehicles.

Evaluation of the unsprung mass effect on ride comfort of in-wheel motor driving vehicles

As an innovative powertrain configuration for electric vehicles, in-wheel motor drive has tremendous potentials. However, the influence of excessive unsprung mass related to the in-wheel motor drive on vehicle performance remains ambiguous or even controversial. This paper tries to address the issue based on a full-vehicle model, taking into consideration different road profiles, body positions and vehicle speeds. A full-vehicle model is established for an in-wheel motor driving vehicle, and a centralized driving vehicle with lighter unsprung mass is taken as the baseline vehicle. Simulation studies are carried out for both vehicle configurations at various speeds on random and bumpy road profiles, respectively. Various observations about the effects of the increased unsprung mass are obtained and clarified, including those mentioned in the literature and some new ones. The results indicate that, for random roads, the increased unsprung mass enlarges the wheel dynamic load and suspension travel, and thus deteriorates the road holding and suspension performances. Interestingly, the effects of the increased unsprung mass on body vertical acceleration vary with body position due to the wheelbase filtering property. Front and rear body accelerations are magnified with the increased unsprung mass at all speeds, whereas the body centroid acceleration exhibits an alternating pattern of increase and decrease throughout the speed range. Moreover, the increased unsprung mass reduces the body roll acceleration at all speeds. For bump roads on the other hand, the increased unsprung mass deteriorates all metrics at low speeds, but improves them at high speeds.

Study of Distribution of Free Flow Speeds on Urban Road Sections Depending on Their Functional Purpose and One-Way Traffic—Evidence from Kharkiv (Ukraine)

Data on the distribution of the free flow speed (FFS) of cars are used to solve a wide range of tasks in the field of road transport, starting from road design and ending with the development of traffic modeling and simulation programs. The purpose of this study was to obtain the distribution of vehicle speeds on typical sections of the city road network, characterized by the presence of one-way traffic. The data were obtained by field observations using a portable radar. As a result, statistical characteristics and speed distribution laws for four sections of streets in the city of Kharkiv were analyzed. It was shown that the characteristics of FFS distributions differ depending on the functional class of the streets. Average FFS values on main street segments were on average 19 km/h higher. The one-way traffic has less impact on the FFS distribution, especially for arterial streets. The characteristics of FFS distributions differ depending on the type and functional class of streets; they can be described with sufficient accuracy by typical distribution laws, such as Normal, Log-normal, Gamma, and Chi-square. The results of this study can be useful for traffic modeling problems.

Extracting Bridge Modal Frequencies Using Stationary Versus Drive-By Modes of Smartphone Measurements

In this research, we harmonize the two mobility approaches, stationary and mobile measurements, within the same framework to generate comparison opportunities, particularly in terms of identified bridge modal frequencies. Vibration tests were conducted to determine the natural frequency of a pedestrian bridge located in University College Dublin using smartphones. Both stationary and mobile smartphone measurements were collected, a novel use of two levels of mobility. Stationary measurements involved leaving the smartphone on the bridge deck at different positions along the bridge for a period of time, and mobile measurements were carried out using an electric scooter to ride across the bridge with the smartphone attached to the scooter deck. Single-output identification results were then compared to visualize the differences at two mobility levels. The tests showed that it is possible to extract the first natural frequency of the bridge using both stationary and mobile smartphone measurement techniques, although operational uncertainties seemed to alter the latter one. A first natural frequency of 5.45 Hz from a reference data acquisition system confirmed the accuracy of stationary smartphone data. On the other hand, the mobile data require consideration of the driving frequency, a function of the speed of the test vehicle and length of the bridge. These results show that smartphone sensors can be regarded as an alternative to industrial accelerometers with certain barriers to account for the multi-modality of the mobile sensing and identification process.

Robust Speed and Spacing Control Framework for Autonomous Vehicles via µ-synthesis with Descriptor Form Representation

Robust Speed and Spacing Control Framework for Autonomous Vehicles via µ-synthesis with Descriptor Form Representation

A Framework for Smart City Traffic Management utilizing BDA and IoT

This paper explores a new approach to traffic flow optimization in smart cities, harnessing the combined power of Big Data Analytics (BDA) and the Internet of Things (IoT). The system utilizes a citywide network of connected sensors to acquire live traffic information, including vehicle speeds, density, and congestion points. These data are thereafter processed applying some top-notch BDA algorithms to identify traffic anomalies and forecast congestion levels, and generate actionable insights. By analyzing this information, the system can dynamically adjust traffic signals, recommend alternative routes, and improve traffic efficiency in real-time. The system's adaptive learning capabilities allow it to continuously enhance its predictions based on new data, ensuring its effectiveness in managing evolving traffic patterns. This intelligent traffic management solution promises to significantly reduce congestion, ameliorate overall mobility and road safety, and contribute to a more sustainable city environment.

Effects of undulating printing paths on axial compressive behaviors of 3D-printed continuous fiber-reinforced multi-cell thin-walled structure

Thin-walled structures are widely used in passive safety systems of the vehicles because of their excellent lightweight and stable progressive deformation behavior. Recently, continuous fiber 3D printing technique has been successfully applied in the manufacturing of thin-walled structures, providing a promising path to meet increasing requirements for energy absorbers as vehicle speed increases. This paper proposed a range of novel undulating printing paths to fabricate two kinds of continuous Kevlar fiber reinforced multi-cell thin-walled structures (CFMSs) with different cross-section shapes, named MT1 and MT2, respectively. For each type of CFMS, three different printing paths (moving back and forth between two layers) were designed to manufacture CFMS with the wall thickness comprised of only a single deposited line. The axial compressive tests were conducted to investigate the compressive response of CFMSs. It was found that undulating printing paths affect both the mechanical behavior and collapse mode of CFMSs. During the compression process, MT1 and MT2 printed with path 3 showed progressively folding deformation modes (differed from the progressive crushing mode of composite laminated thin-walled structures). Thus, compared to MT2 printed with path 1 and path 2, which exhibit global buckling, MT1 and MT2 printed with path 3 exhibited more stable energy absorption performance. Besides, although the local failure mode in different CFMSs caused by the same path was similar, the same printing path had different influences on the axial compressive properties of different CFMSs.

Assumptions for the mathematical model investigating the effect of railway line route parameters on the average speed of railway vehicles

Rail transport is the most efficient alternative to road transport. The popularity of this branch is constantly growing. As a consequence, it is noticed that key terminals, railway routes and HUB stations are constantly being overloaded, resulting in a reduction in the available capacity. This necessitates the passage of higher priority trains (mainly passenger trains), which requires the introduction of additional stops for freight trains and consequently reduces commercial speeds. A railway line is characterized, m.in, by parameters such as: electrification, maximum speed, railway line class, track width, etc. Each of these parameters can have an impact on the average speed of rail vehicles in rail transport. An additional factor determining speed is the capacity of railway lines. Due to the lower capacity and the consequent low average commercial speed, alternative routes for trains are being sought. The aim of the article is to analyze the assumptions for the mathematical model of the influence of technical parameters of routes and capacity of railway lines on the average speed of rail vehicles. In order to achieve this goal, the article is divided into five sections. The first section is an introduction to the article. The second section describes theoretical issues related to modelling in transport. The third section describes the issues of mathematical modeling using graph theory. The practical application of graph theory on the selected example is described in the fourth section. A summary of the article and preliminary research is presented in the last section.

DEVELOPING A CONTROLLER FOR AN ADAPTIVE CRUISE CONTROL (ACC) SYSTEM: UTILIZING DEEP REINFORCEMENT LEARNING (DRL) APPROACH

The addition of Adaptive Cruise Control (ACC) to vehicles enables automatic speed adjustments based on traffic conditions after the driver sets the maximum speed, freeing them to concentrate on steering. This study is dedicated to the development of a passenger car ACC system using Deep Reinforcement Learning (DRL). A critical aspect of this ACC system is its capability to regulate the distance between vehicles by taking into account preceding and following vehicle speeds. It considers three primary inputs: the memory-stored speed of the following vehicle, the lead time specified by the driver, and the radar-measured distance. By adapting speed in different traffic scenarios, the system contributes to averting potential accidents. This research delves into constructing a controller that utilizes the Deep Deterministic Policy Gradient (DDPG) algorithm and compares its outcomes with those from the DQN algorithm. The DDPG controller supervises the longitudinal control actions of a vehicle, enabling it to execute stopping and moving maneuvers safely and efficiently.

Research on Risk Quantification Methods for Connected Autonomous Vehicles Based on CNN-LSTM

Quantifying and predicting driving risks for connected autonomous vehicles (CAVs) is critical to ensuring the safe operation of traffic in complex environments. This study first establishes a car-following model for CAVs based on molecular force fields. Subsequently, using a convolutional neural network and long short-term Memory (CNN-LSTM) deep-learning model, the future trajectory of the target vehicle is predicted. Risk is quantified by employing models that assess both the collision probability and collision severity, with deep-learning techniques applied for risk classification. Finally, the High-D dataset is used to predict the vehicle trajectory, from which the speed and acceleration of a target vehicle are derived to forecast driving risks. The results indicate that the CNN-LSTM model, when compared with standalone CNN and LSTM models, demonstrates a superior generalization performance, a higher sensitivity to risk changes, and an accuracy rate exceeding 86% for medium- and high-risk predictions. This improved accuracy and efficacy contribute to enhancing the overall safety of connected vehicle platoons.

Experimental assessment of communication delay's impact on connected automated vehicle speed volatility and energy consumption

Communication delays within connected and autonomous vehicles (CAVs) pose significant risks. It is imperative to address these issues to ensure the safe and effective operation of CAVs. However, the exploration of communication delays on CAV operations and their energy use remains sparse in the literature. To fill the research gap, this study leverages the facilities at America Center of Mobility (ACM) Smart City Test Center to implement and evaluate a CAV merging control algorithm through vehicle-in-the-loop testing. This study aims at achieving three main objectives: (1) develop and implement a CAV merging control strategy in the experimental test bed through vehicle-in-the-loop testing, (2) propose analytical models to quantify the impacts of communication delay on the variability of CAV speed and energy consumption based on field experiment data, and (3) create a predictive model for energy usage considering various CAV attributes and dynamics, e.g., speed, acceleration, yaw rate, and communication delays. To our knowledge, this is one of the first attempts at evaluating the impacts of communication delays on CAV merging operational control with field data, making critical advancement in the field. The results suggest that communication delay has a more substantial effect on energy consumption under high-speed volatility compared to low-speed volatility. Among all factors examined, acceleration is the dominant characteristic that influences energy usage. It also revealed that even minor improvements in communication delay can yield tangible improvements in energy efficiency. The results provide guidance on CAV field experiments and the influence of communication delays on CAV operation and energy consumption.

A Single-Stage SOC Reference Trajectory Prediction Method for Series PHEV Based on GRNN

In this paper, a single-stage SOC reference trajectory (SOC-RT) prediction method is proposed. Different from the two-stage prediction method mostly used in previous studies, the proposed strategy predicts the SOC-RT slope directly based on the operating condition, instead of predicting the vehicle speed first and then solving it based on a global optimization algorithm. Therefore, the proposed method can significantly reduce the computational load and has high application value. A generalized regression neural network (GRNN) is employed as the prediction model for its strong nonlinear fitting ability. Meanwhile, to solve the problem of error accumulation due to the slope prediction, the prediction trajectory is corrected by the linear trajectory, and the correction factor is determined based on particle swarm optimization algorithm (PSO). Finally, the trajectory is followed by equivalent consumption minimization strategy (ECMS) and the performance of the proposed strategy is verified. The results show that the prediction accuracy is improved by 52.10% and 3.73% compared to the linear trajectory under the two tested cycles, respectively. Meanwhile, the proposed strategy achieves the best fuel economy, with fuel consumption decreasing by 2.61% and 6.40%, respectively, compared to the rule-based control strategy. Finally, its real-time control performance in a real hardware environment is verified through the hardware-in-the-loop test.

Modeling and Simulation of Vehicle Velocity-Density on Buah Batu Road Using Second-Order Polynomial Regression

The problem of traffic density is a complex problem in the world of land transportation, especially in urban areas, including Bandung City. Buah Batu Road, one of the main roads in Bandung City which is 13 meters wide and 1.70 kilometers long, connecting Bandung City and Bandung Regency. And this study examines the relationship between vehicle speed and traffic density on Buah Batu Road, Bandung. Using the macroscopic Lighthill-Whitham Richards (LWR) model, Second Order Polynomial Regression, and Lax-Wendroff scheme simulation. This study aims to obtain the speed-density function for traffic. The introduction emphasizes the importance of understanding traffic flow dynamics to reduce congestion, especially in areas with significant vehicle growth. The methodology used is direct observation of the Buah Batu Road section with an observed length of 18 meters, with data collected through cellphone camera recordings at various times. These observation data provide insight into vehicle density and speed under various conditions.

FPGA-Based Design for Digital Speedometer to Detect Speed Limit of Vehicles

This paper presents a proposed design for a digital speedometer using Field Programmable Gate Array (FPGA) technology. The speedometer is a crucial safety feature in vehicles, providing real-time information on the vehicle's speed in kilometres per hour (km/h). The objective of this design is to assist drivers in monitoring and controlling their velocity at regular intervals to ensure safe driving. The hardware platform chosen for implementation is the Altera DE2 Board with Cyclone IV E and the hardware description language used is Verilog HDL. The digital speedometer receives a square wave pulse as an input signal from a hall sensor, which is emitted based on the vehicle's speed. The designed system accurately detects and counts these pulses to convert them into the corresponding speed of the vehicle. This digital speedometer is specifically intended for vehicles with engine capacities (cc) ranging from 1000cc to 1300cc. It can detect speeds within the range of 0km/h to 255km/h. The speed value is displayed on a seven-segment display with precision. Additionally, if the vehicle's speed exceeds the expressway's speed limit of 110km/h, a red LED indicator turns on to alert the driver to control their speed.

Analysis of the Implementation of Traffic Management at the Johor Roundabout Jalan Karya Wisata-Medan Johor

Johor Roundabout is located on Jalan Karyawisata kec, Medan Johor, or more precisely right in front of the entrance to the J-City complex. This roundabout was just built in the 12th month of 2023. The existence of this roundabout has a detrimental impact on the speed of vehicles passing through the roundabout. The aim of this thesis is to analyze the performance of roundabouts using the MKJI1997 method. This survey was carried out by conducting research for 7 days, where each day I conducted a survey for 12 hours, namely from 07:00 to 19:00. After carrying out the survey, it was also found that the peak hour occurred on Monday at 17:00-18:00 where it was obtained that the total inflow was 7854 PCU/hour with the composition of Light Vehicles (LV) 3127 Pc/hour, Heavy Vehicles (HV) 322 PCU/hour and , motorbike (MC) 2189Smp/hour, and the results also obtained for the degree of saturation, namely 1.094, which means that the level of service for the roundabout is at F (i.e. traffic flow conditions are relatively low).

Pedestrian Danger Index – a novel surrogate measure of pedestrian safety at unsignalized crossings

Objective The aims of the study are to assess pedestrian safety at unsignalized crossings and to develop a new surrogate measure to evaluate risks associated with vehicle-pedestrian interactions. The primary goal of developing a novel safety metric was to provide quick insights into pedestrian safety levels and conduct site assessments rather than predictive analyses. Methods Video surveys were conducted at multiple unsignalized crossing sites in Poland to record vehicle-pedestrian interactions. Vehicle and pedestrian trajectories were extracted using video image analysis. An interaction classification algorithm was introduced, based on relative movements of the two road users involved. Using selected video recorded interactions, an expert opinion survey was conducted in order to evaluate pedestrians’ perceptions of risk. The level of risk was modeled with a "danger degree function", incorporating passing distance, vehicle speed, and deceleration. Results Analyses revealed that smaller passing distance and higher vehicle speeds are linked to increased danger levels. Pedestrian Danger Index (PDI) was introduced as a new measure to assess pedestrian safety by averaging danger levels for all the observed pedestrian-vehicle interactions. Based on the sensitivity analysis, it is recommended to use a minimum of 4 survey days when calculating the PDI indicator. Conclusions The PDI offers a novel method for evaluating pedestrian safety at unsignalized crossings, though it has some limitations. This user-friendly surrogate safety measure relies on three key variables for computation: passing distance, vehicle speed, and vehicle deceleration rates. The formulated PDI equation can be used for analyzing pedestrian crossings without traffic signals. Further research is recommended to refine the PDI and enhance its effectiveness.

Effects of Lane Imbalance on Capacity Drop and Emission in Expressway Merging Areas: A Simulation Analysis

Lane imbalance does not provide sufficient space for merging vehicles to adjust their speed and change lanes smoothly. This leads to improper driving behavior that disrupts mainline traffic flow stability, resulting in capacity drops and increased vehicle emissions. However, quantitative analyses, specifically the effects of lane imbalance on capacity and emissions, remain limited. Existing traffic simulation platforms struggle to capture the effects of geometric design changes on capacity. To address these gaps, we developed a simulation method incorporating interactions between geometric design and traffic flow demand into an XGBoost model, enhancing the predictive accuracy for driving behavior parameters. Implemented within the TESS NG platform, this model enables real-time adjustments in driving behavior parameters as traffic demand varies under different lane balance conditions. The simulation results indicated a 42.4% capacity drop and a 34.9% increase in CO2 emissions when the balanced merging area was shifted to lane imbalance. Conversely, shifting to lane balance increases capacity by 8.2% and reduces CO2 emissions by 39.8% under severe congestion conditions. Under lane imbalance, vehicle speeds are lower across all traffic demand levels. When the demand exceeds 1300 pcu/h/ln, lane changes occur closer to the end of the acceleration lane, with higher speed differentials. These insights underscore the potential of lane balance optimization to mitigate capacity drops and emissions, providing a valuable simulation approach for the design and evaluation of merging areas.