Browse Topic: Traffic management

Items (498)
ABSTRACT This work presents the development of an algorithm to incorporate measurements from multiple antennas to improve the relative position solution between convoying vehicles provided by Global Positioning System (GPS) measurements. The technique presented, incorporates measurements from multiple antennas with a known fixed-baseline between a base antenna and auxiliary antenna on a base vehicle, and a rover antenna on a rover vehicle. The additional information provided by the fixed-baseline distance is used to provide an additional measurement with low uncertainty for improved integer ambiguity resolution between the base and auxiliary receiver, which in turn, provides additional measurements for determining the integer ambiguity difference between the base and rover receivers for the computation of a high-precision relative position vector (HPRPV
Tabb, Thomas T.Bevly, DavidMartin, ScottRatowski, Jeff
ABSTRACT Military personnel involved in convoy operations are often required to complete multiple tasks within tightly constrained timeframes, based on limited or time-sensitive information. Current simulations are often lacking in fidelity with regard to team interaction and automated agent behavior; particularly problematic areas include responses to obstacles, threats, and other changes in conditions. More flexible simulations are needed to support decision making and train military personnel to adapt to the dynamic environments in which convoys regularly operate. A hierarchical task analysis approach is currently being used to identify and describe the many tasks required for effective convoy operations. The task decomposition resulting from the task analysis provides greater opportunity for determining decision points and potential errors. The results of the task analysis will provide guidance for the development of more targeted simulations for training and model evaluation from
Garrison, Teena M.Thomas, Mark D.Carruth, Daniel W.
ABSTRACT Occupant safety is a top priority of military vehicle designers. Recent trends have shifted safety emphasis from the threats of ballistics and missiles toward those of underbody explosives. For example, the MRAP vehicle is increasingly replacing the HMMWV, but it is much heavier and consumes twice as much fuel as its predecessor. Recent reports have shown that fuel consumption directly impacts personnel safety; a significant percentage of fuel convoys that supply current field operations experience casualties en route. While heavier vehicles tend to fare better for safety in blast situations, they contribute to casualties elsewhere by requiring more fuel convoys. This study develops an optimization framework that uses physics-based simulations of vehicle blast events and empirical fuel consumption data to calculate and minimize combined total expected injuries from blast events and fuel convoys. Results are presented by means of two parametric studies, and the utility of the
Hoffenson, StevenKokkolaras, MichaelPapalambros, PanosArepally, Sudhakar
ABSTRACT To improve robustness of autonomous vehicles, deployments have evolved from a single intelligent system to a combination of several within a platoon. Platooning vehicles move together as a unit, communicating with each other to navigate the changing environment safely. While the technology is robust, there is a large dependence on data collection and communication. Issues with sensors or communication systems can cause significant problems for the system. There are several uncertainties that impact a system’s fidelity. Small errors in data accuracy can lead to system failure under certain circumstances. We define stale data as a perturbation within a system that causes it to repetitively rely on old data from external data sources (e.g. other cars in the platoon). This paper conducts a fault injection campaign to analyze the impact of stale data in a platooning model, where stale data occurs in the car’s communication and/or perception system. The fault injection campaign
Louis, August St.Calhoun, Jon C.
ABSTRACT Leader-follower autonomous vehicle systems have a vast range of applications which can increase efficiency, reliability, and safety by only requiring one manned-vehicle to lead a fleet of unmanned followers. The proper estimation and duplication of a manned-vehicle’s path is a critical component of the ongoing development of convoying systems. Auburn University’s GAVLAB has developed a UWB-ranging based leader-follower GNC system which does not require an external GPS reference or communication between the vehicles in the convoy. Experimental results have shown path-duplication accuracy between 1-5 meters for following distances of 10 to 50 meters. Citation: K. Thompson, B. Jones, S. Martin, and D. Bevly, “GPS-Independent Autonomous Vehicle Convoying with UWB Ranging and Vehicle Models,” In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI, Aug. 16-18, 2022
Thompson, KyleJones, BenMartin, ScottBevly, David
ABSTRACT This paper will document the development of the Convoy Active Safety Technology (CAST) program, which was created to design a low cost, optionally manned vehicle (OMV) solution for tactical wheeled vehicle (TWV) fleet. This paper will describe the approach taken to integrate low cost sensors for understanding the environment sufficiently to accomplish convoy missions. This paper will also discuss the approach taken to develop the low cost guidance and navigation solution used in the CAST program
Simon, DavidTheisen, Bernard
ABSTRACT As unmanned ground vehicle technology matures and autonomous platforms become more common, such platforms will invariably be in close proximity to one another both in formation and independently. With an increasingly crowded field, the risk of collisions between these platforms grows, and with it the need for path deconfliction. This paper presents two complementary technological developments to this end: a pipeline for affirmatively identifying and classifying dynamic objects, e.g., vehicles or pedestrians; and a pipeline for preventing collisions with such objects. The efficacy of these techniques is demonstrated in simulation, and validation on robotic platforms will be undertaken in the near future. Citation: Matthew Grogan, “Dynamic Object Collision Avoidance for Autonomous Multi-Vehicle Systems in the Robotic Technology Kernel”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI, Aug. 13-15, 2019
Grogan, Matthew
The highway diverging area is a crucial zone for highway traffic management. This study proposes an evaluation method for traffic flow operations in the diverging area within an Intelligent and Connected Environment (ICE), where the application of Connected and Automated Vehicles (CAVs) provides essential technical support. The diverging area is first divided into three road sections, and a discrete state transition model is constructed based on the discrete dynamic traffic flow model of these sections to represent traffic flow operations in the diverging area under ICE conditions. Next, an evaluation method for the self-organization degree of traffic flow is developed using the Extended Entropy Chaos Degree (EECD) and the discrete state transition model. Utilizing this evaluation method and the Deep Q-Network (DQN) algorithm, a short-term vehicle behavior optimization method is proposed, which, when applied continuously, leads to a vehicle trajectory optimization method for the
Fang, ZhaodongQian, PinzhengSu, KaichunQian, YuLeng, XiqiaoZhang, Jian
ABSTRACT The application of advanced FEV Automotive Smart Vehicle© methods and technologies while maintaining functional safety compliance and how it applies to similar features, requirements and capabilities across the fleet of DoD combat and tactical vehicles will be discussed. The requirement of technologies for DoD autonomous ground vehicle including leader follower, automated convoy operations, and intelligent applique kit’ are common to those specified in the automotive industries. Intelligent vehicles can be advanced and implemented in an expeditious manner through FEV Smart Vehicle technologies, techniques and methodologies while maintain compliance to required functional safety. The application and impact of ISO 26262 (2011) as well as Mil-Std. 882(E) to the implementation of the advanced technologies and techniques in support of full operational vehicle autonomy can hinder development. Leveraging the FEV Automotive Smart Vehicle reduces the time and cost for safety compliant
LaRue, David A.Tasky, TomTarnutzer, StephanLane, Jerry
ABSTRACT The automotive and defense industries are going through a period of disruption with the advent of Connected and Automated Vehicles (CAV) driven primarily by innovations in affordable sensor technologies, drive-by-wire systems, and Artificial Intelligence-based decision support systems. One of the primary tools in the testing and validation of these systems is a comparison between virtual and physical-based simulations, which provides a low-cost, systems-approach testing of frequently occurring driving scenarios such as vehicle platooning and edge cases and sensor-spoofing in congested areas. Consequently, the project team developed a robotic vehicle platform—Scaled Testbed for Automated and Robotic Systems (STARS)—to be used for accelerated testing elements of Automated Driving Systems (ADS) including data acquisition through sensor-fusion practices typically observed in the field of robotics. This paper will highlight the implementation of STARS as a scaled testbed for rapid
Lodato, DiegoKamalanathsharma, RajFarber, Maurice
ABSTRACT Sharing information among vehicles in an unmanned ground vehicle (UGV) convoy allows for improved vehicle performance and reduces the need for each vehicle to be equipped with a full-suite of sensors. Information such as obstacle data, surface properties, and terrain maps are particularly useful for vehicle control and high-level behaviors. This paper describes a system architecture for sharing semantic information among vehicles in a convoy operation. This architecture is demonstrated by sharing terrain information between vehicles in a two-vehicle convoy in both simulation and on actual autonomous vehicles. Update rules fuse information from different sources in a statistical manner and allow for an onboard algorithm to make high-level decisions about the incoming data whether it be from its own sensors or semantic information from other vehicles
Ferrin, Jeffrey L.Bybee, Taylor C.
ABSTRACT This paper presents Neya’s efforts in developing autonomous depot assembly and parking behaviors for the Ground Vehicle Systems Center’s (GVSC) Autonomous Ground Re-supply (AGR) program. Convoys are a prime target for the enemy, and therefore GVSC is making efforts to remove the human operators and make them autonomous. However, humans still have to manually drive multiple convoy vehicles to and from their depot parking locations before and after autonomous convoy operations – a time-consuming and laborious process. Neya systems was responsible for the design, development, and testing of the autonomous depot assembly and disassembly behaviors, enabling end-to-end autonomy for convoy operations. Our solution to the problem, including the concept of operations, design, as well as approaches towards testing and validation are described in detail
Mattes, RichBruck, KurtCascone, AnthonyMartin, Dave
ABSTRACT In this paper, we present CLICS, a program that optimizes convoy vehicle tracks by intelligently combining sensor updates of all vehicles in the convoy in a distributed, cooperative localization system. Currently, follower vehicles in the convoy rely either on GPS breadcrumbs from the lead vehicle, or rely on sensing the location of its predecessor and following its path. However, GPS availability and accuracy oftentimes cause the former solution to fail, and accumulated errors in tracking and control in long convoys can cause the latter solution to fail. Robotic Research’s CLICS system attempts to overcome these problems by (1) integrating multiple heterogeneous sensor outputs from multiple vehicles (2) developing a distributed, real-time non-linear estimation of inter-vehicle pose using spring network providing coordinated localization for members of a vehicle convoy, and (3) real-time robust synchronization of information amongst the convoy, and local convoy and mission
Wilhelm, RayBalas, CristianSchneider, AnneKlarquist, WilliamLacaze, AlbertoMurphy, Karl
ABSTRACT In this paper, we will present the results of our efforts developing the Autonomy Kit for the Tank Automotive Research Development and Engineering Center’s (TARDEC) Autonomous Ground Resupply (AGR) Sustainment Operations (SO) program. Robotic Research, LLC was responsible for the design, build, and implementation of the “Autonomy Kit” for the AGR SO. The Autonomy Kit is designed to be a fault-tolerant, vehicle-agnostic applique kit that provides the hardware and software needed to perform higher-level autonomous driving and planning functions. In the first Increment, the main focus was developing a “Leader/Follower” capability, where a manned “Leader” vehicle could perform a mission with a number of unmanned “Followers” reproducing its trajectory, maintaining convoy constraints, and avoiding obstacles in the path
Schneider, AnneBecker, WilliamHoward, JohnCichosz, AlanTheisen, BernardConger, David
ABSTRACT Many significant advances have been made in autonomous vehicle technology over the recent decades. This includes platooning of heavy trucks. As such, many institutions have created their own version of the basic platooning platform. This includes the California PATH program [1], Japan’s “Energy ITS” project [2], and Auburn University’sCACC Platform [3]. One thing these platforms have in common is a strong dependence on GPS based localization solutions. Issues arise when the platoon navigates into challenging environments, including rural areas with foliage which might block receptions, or more populated areas which might present urban canyon effects. Recent research focus has shifted to handling these situations through the use of alternative sensors, including cameras. The perception method proposed in this paper utilizes the You Only Look Once (YOLO) real-time object detection algorithm in order to bound the lead vehicle using both RGB and IR cameras. Range and bearing are
Flegel, TylerChen, HowardBevly, David
ABSTRACT We describe a simulation environment that enables the design and testing of control policies for off-road mobility of autonomous agents. The environment is demonstrated in conjunction with the design and assessment of a reinforcement learning policy that uses sensor fusion and inter-agent communication to enable the movement of mixed convoys of conventional and autonomous vehicles. Policies learned on rigid terrain are shown to transfer to hard (silt-like) and soft (snow-like) deformable terrains. The enabling simulation environment, which is Chrono-centric, is used as follows: the training occurs in the GymChrono learning environment using PyChrono, the Python interface to Chrono. The GymChrono-generated policy is subsequently deployed for testing in SynChrono, a scalable, cluster-deployable multi-agent testing infrastructure that uses MPI. The Chrono::Sensor module simulates sensing channels used in the learning and inference processes. The software stack described is open
Negrut, D.Serban, R.Elmquist, A.Taves, J.Young, A.Tasora, A.Benatti, S.
ABSTRACT This paper presents two techniques for autonomous convoy operations, one based on the Ranger localization system and the other a path planning technique within the Robotic Technology Kernel called Vaquerito. The first solution, Ranger, is a high-precision localization system developed by Southwest Research Institute® (SwRI®) that uses an inexpensive downward-facing camera and a simple lighting and electronics package. It is easily integrated onto vehicle platforms of almost any size, making it ideal for heterogeneous convoys. The second solution, Vaquerito, is a human-centered path planning technique that takes a hand-drawn map of a route and matches it to the perceived environment in real time to follow a route known to the operator, but not to the vehicle. Citation: N. Alton, M. Bries, J. Hernandez, “Autonomous Convoy Operations in the Robotic Technology Kernel (RTK)”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI
Alton, NicholasBries, MatthewHernandez, Joseph
In recent times there has been an upward trend in “Connected Vehicles”, which has significantly improved not only the driving experience but also the “ownership of the car”. The use of state-of-the-art wireless technologies, such as vehicle-to-everything (V2X) connectivity, is crucial for its dependability and safety. V2X also effectively extends the information flow between the transportation ecosystem pedestrians, public infrastructure (traffic management system) and parking infrastructure, charging and fuel stations, Etc. V2X has a lot of potential to enhance traffic flow, boost traffic safety, and provide drivers and operators with new services. One of the fundamental issues is maintaining trustworthy and quick communication between cars and infrastructure. While establishing stable connectivity, reducing interference, and controlling the fluctuating quality of wireless transmissions, we have to ensure the Security and Privacy of V2I. Since there are multiple and diverse
Sundar, ShyamPundalik, KrantiveerUnnikrishnan, Ushma
Background: Road accident severity estimation is a critical aspect of road safety analysis and traffic management. Accurate severity estimation contributes to the formulation of effective road safety policies. Knowledge of the potential consequences of certain behaviors or conditions can contribute to safer driving practices. Identifying patterns of high-severity accidents allows for targeted improvements in terms of overall road safety. Objective: This study focuses on analyzing road accidents by utilizing real data, i.e., US road accidents open database called “CRSS.” It employs advanced machine learning models such as boosting algorithms such as LGBM, XGBoost, and CatBoost to predict accident severity classification based on various parameters. The study also aims to contribute to road safety by providing predictive insights for stakeholders, functional safety engineering community, and policymakers using KABCO classification systems. The article includes sections covering
Babaev, IslamMozolin, IgorGarikapati, Divya
Artificial Intelligence (AI) has emerged as a transformative force across various industries, revolutionizing processes and enhancing efficiency. In the automotive domain, AI's adaption has ushered in a new era of innovation and driving advancements across manufacturing, safety, and user experience. By leveraging AI technologies, the automotive industry is undergoing a significant transformation that is reshaping the way vehicles are manufactured, operated, and experienced. The benefits of AI-powered vehicles are not limited to their manufacturing, operation, and enhancing the user experience but also by integrating AI-powered vehicles with smart city infrastructure can unlock much more potential of the technology and can offer numerous advantages such as enhanced safety, efficiency, growth, and sustainability. Smart cities aim to create more livable, resilient, and inclusive communities by harnessing innovation through technologies like Internet of Things (IoT), devices, data
Shrimal, Harsh
Eco-driving algorithms use the available information about traffic and route conditions to optimize the vehicle speed and achieve enhanced energy consumption while fulfilling a travel time constraint. Depending on what information is available, when it becomes accessible, and the level of automation of the vehicle, different energy savings can be achieved. In their basic formulation, eco-driving algorithms only leverage static information to evaluate the optimal speed, such as posted speed limits and location of stop signs. More advanced algorithms may also consider dynamic information, such as the speed of the preceding vehicle and Signal Phase and Timing of traffic lights, thus achieving higher energy efficiency. The objective of the proposed work is to develop an eco-driving algorithm that can optimize energy consumption by leveraging not only static route information, but also dynamic macroscopic traffic conditions, which are assumed to be available in real-time through
Villani, ManfrediShiledar, AnkurBlock, BrianSpano, MatteoRizzoni, Giorgio
Traditional live testing of autonomous ground vehicles can be augmented through use of digital twins of the test environment, the vehicle mobility models, and the vehicle sensors. These digital twins combined with the autonomous software under test allow testers to inject faults, weather, obstacles, find edge case scenarios, and collect information to understand the decision making of the autonomous software under test. With this new capability, autonomous ground vehicles can now be tested in four stages. The first stage is testing the autonomous software using digital twins. In this stage with the help of a High-Performance Computer thousands of scenarios can be run. Once issues are communicated and addressed, stage two, hardware in the loop testing can begin. Hardware in the loop uses simulators that already exist to test systems such as autonomous convoys with a virtual leader and a live follower. Stage three employs a live virtual constructive approach by using one vehicle to test
Whitt, John M.Bounker, Paul J.
The deployment of autonomous urban buses brings with it the hope of addressing concerns associated with safety and aging drivers. However, issues related autonomous vehicle (AV) positioning and interactions with road users pose challenges to realizing these benefits. This report covers unsettled issues and potential solutions related to the operation of autonomous urban buses, including the crucial need for all-weather localization capabilities to ensure reliable navigation in diverse environmental conditions. Additionally, minimizing the gap between AVs and platforms during designated parking requires precise localization. Next-gen Urban Buses: Autonomy and Connectivity addresses the challenge of predicting the intentions of pedestrians, vehicles, and obstacles for appropriate responses, the detection of traffic police gestures to ensure compliance with traffic signals, and the optimization of traffic performance through urban platooning—including the need for advanced communication
Hsu, Tsung-Ming
This article offers an algorithmic solution for moving a homogeneous platoon of position-controlled vehicles on a curved path with varying speeds and in the presence of communication losses and delays. This article considers a trajectory-based platooning with the leader–following communication topology, where the lead vehicle communicates its reference position and orientation to each autonomous follower vehicle. A follower vehicle stores this communicated information for a specific period as a virtual trail of the lead vehicle starting from the lead vehicle’s initial position and orientation. An algorithm uses this trail to find the follower vehicle’s reference position and orientation on that trail, such that the follower vehicle maintains a constant distance from the lead vehicle. The proposed algorithm helps form a platoon where each vehicle can traverse a curve with varying speeds. In contrast, in the existing literature, most of the solutions for vehicle platooning on a curved
Bhaskar, RintuWahi, PankajPotluri, Ramprasad
There have been numerous studies on stable platooning, but almost all of them have been on the longitudinal stability problem, wherein, without sufficient longitudinal stability, traffic congestion might occur more frequently than in traffic consisting of manually driven vehicles. Failure to solve this problem would reduce the value of autonomous driving. Recently, some researchers have begun to tackle the lateral stability problem, anticipating shortened intervehicle distances in the future. Here, the intervehicle distance in a platoon should be shortened to improve transportation efficiency. However, if an obstacle to be avoided exists, the following vehicles might have difficulty finding it quickly enough if the preceding vehicle occludes it from their sensors. Also, longer platoons improve transportation efficiency because the number of gaps between platoons is reduced. Hence, in this study, the lateral stability of platoons consisting of autonomous vehicles was analyzed for not
Kurishige, Masahiko
This research investigates platoon dispersion characteristics in mixed-traffic flow of autonomous and human-driven vehicles. It presents a cellular automata-based platoon dispersion model. The study’s key findings are as follows: platoon dispersion initially increases and then decreases with the rise in autonomous vehicle proportions. When the autonomous vehicle proportion is approaching 100%, platoon dispersion descends rapidly and is completely eliminated while the proportion is 100%. Compared to platoon consisting entirely of human-driven vehicles, the peak value of standard deviation of vehicle speed is 1.71 times and the travel time drops by 38.19% when the proportion is 1. Moreover, the lane-changing behavior enhances platoon speed, acceleration, and space utilization at micro- and macrolevels by optimizing space resource allocation within the platoon. The study employs a two-lane mixed-flow platoon dispersion model that assumes uniform vehicle characteristics and prioritizes
Lu, TingLiu, ChenghaoLin, SitongSong, Wenjing
Truck platooning facilitates the operation of trucks in close proximity to one another, resulting in decreased air resistance and improved fuel efficiency. While previous research has mostly focused on the effects of intra-distance on fuel savings, this study aims to develop fuel savings performance functions considering various truck platooning configurations. This article comprehensively investigates the influence of different truck platoon configurations on fuel savings. This analysis focuses on examining the impacts of several variables including inter-vehicle distance, platoon speed, truck weight, number of trucks in the platoon, and the truck’s distinctive design characteristics. Data used in the analysis were collected from 10 different field experiments. Three machine learning techniques—artificial neural networks (ANN), extreme gradient boosting (XGBoost), and K-nearest neighbors (KNN)—alongside the negative binomial regression model were employed. Upon evaluation, the
Mohamed, MohamedHassan, Hany M.
Urban Air Mobility (UAM) envisions heterogenous airborne entities like crewed and uncrewed passenger and cargo vehicles within, and between urban and rural environment. To achieve this, a paradigm shift to a cooperative operating environment similar to Extensible Traffic Management (xTM) is needed. This requires the blending of traditional Air Traffic Services (ATS) with the new generation UAM vehicles having their unique flight dynamics and handling characteristics. A hybrid environment needs to be established with enhanced shared situational awareness for all stakeholders, enabling equitable airspace access, minimizing risk, optimized airspace use, and providing flexible and adaptable airspace rules. This paper introduces a novel concept of distributed airspace management which would be apt for all kinds of operational scenarios perceived for UAM. The proposal is centered around the efficiency and safety in air space management being achieved by self-discipline. It utilizes
KG, SreenivasanSuseelan, SunilRajHuncha, Pradeep
With the rapid growth of automobile ownership, traffic congestion has become a major concern at intersections. In order to alleviate the blockage of intersection traffic flow caused by signals, reduce the length of vehicle congestion and waiting time, and for improving the intersection access efficiency, therefore, this article proposes a vehicle speed guidance strategy based on the intersection signal change by combining the vehicle–road cooperative technology. The randomness of vehicle traveling speed in the road is being considered. According to the vehicle traveling speed, a speed guidance model is established under different conditions. Finally, the effectiveness of the speed guidance strategy in this article is verified through experimental simulation, and the benefits of the intersection with intelligent control and traditional control are compared, and the experimental results show that the intelligent control method in this article can effectively reduce vehicle congestion and
Li, WenliLi, AnRen, YongpengWang, Kan
Truck platooning is an emerging technology that exploits the drag reduction experienced by bluff bodies moving together in close longitudinal proximity. The drag-reduction phenomenon is produced via two mechanisms: wake-effect drag reduction from leading vehicles, whereby a following vehicle operates in a region of lower apparent wind speed, thus reducing its drag; and base-drag reduction from following vehicles, whereby the high-pressure field forward of a closely-following vehicle will increase the base pressure of a leading vehicle, thus reducing its drag. This paper presents a physics-guided empirical model for calculating the drag-reduction benefits from truck platooning. The model provides a general framework from which the drag reduction of any vehicle in a heterogeneous truck platoon can be calculated, based on its isolated-vehicle drag-coefficient performance and limited geometric considerations. The model is adapted from others that predict the influence of inter-vehicle
McAuliffe, Brian
Autonomous Vehicles are being widely tested under diverse conditions with expectations that they will soon be a regular feature on roads. The development of Autonomous Vehicles has become an important policy in countries around the world, and the technologies developed by countries and car manufacturers are different, and at the same time to adapt to the road environment and traffic management facilities of different countries, so some countries have built self-driving test sites, and the test content is also different, so it is impossible to compare its relative difficulty. This study surveyed experts and scholars to develop a means of weighting the respective difficulty of various autonomous vehicle testing conditions based on the analytic hierarchy process and fuzzy analytic hierarchy process, applied to a sample of 33 sets of testing conditions based on road type, management actions and operational capabilities. Weights are also adjusted in response to environmental impact factors
Lin, Da-JieLiu, Hsin HsienCHOU, AI-CHENHuang, Pin-ChengWU, CHENG HSINCHANG, CHUN-YICHEN, MING-HSU
As a key technology of intelligent transportation system, vehicle type recognition plays an important role in ensuring traffic safety,optimizing traffic management and improving traffic efficiency, which provides strong support for the development of modern society and the intelligent construction of traffic system. Aiming at the problems of large number of parameters, low detection efficiency and poor real-time performance in existing vehicle type recognition algorithms, this paper proposes an improved vehicle type recognition algorithm based on YOLOv5. Firstly, the lightweight network model MobileNet-V3 is used to replace the backbone feature extraction network CSPDarknet53 of the YOLOv5 model. The parameter quantity and computational complexity of the model are greatly reduced by replacing the standard convolution with the depthwise separable convolution, and enabled the model to maintain higher accuracy while having faster reasoning speed. Secondly, the attention mechanism in
Liu, XinHong
With economic development and the increasing number of vehicles in cities, urban transport systems have become an important issue in urban development. Efficient traffic signal control is a key part of achieving intelligent transport. Reinforcement learning methods show great potential in solving complex traffic signal control problems with multidimensional states and actions. Most of the existing work has applied reinforcement learning algorithms to intelligently control traffic signals. In this paper, we investigate the agent-based reinforcement learning approach for the intelligent control of ramp entrances and exits of urban arterial roads, and propose the Proximal Policy Optimization (PPO) algorithm for traffic signal control. We compare the method controlled by the improved PPO algorithm with the no-control method. The simulation experiments used the open-source simulator SUMO, and the results showed that the reinforcement learning control ramp technique increases the average
Ouyang, ChenZhan, ZhenfeiQian, LiuzhuZou, Jie
Platooning is a coordinated driving strategy by which following trucks are placed into the wake of leading vehicles. Doing this leads to two primary benefits. First, the vehicles following are shielded from aerodynamic drag by a “pulling” effect. Secondly, by placing vehicles behind the leading truck, the leading vehicles experience a “pushing” effect. The reduction in aerodynamic drag leads to reduced fuel usage and, consequently, reduced greenhouse gas emissions. To maximize these effects, the inter-vehicle distance, or headway, needs to be minimized. In current platooning strategy iterations, Coordinated Adaptive Cruise Control (CACC) is used to maintain close following distances. Many of these strategies utilize the fuel rate signal as a controller cost function parameter. By using fuel rate, current control strategies have limited applicability to non-conventional powertrains. Vehicle Specific Power (VSP) has shown promise as a metric by which the performance of such controllers
Bentley, JohnStegner, EvanBevly, David M.Hoffman, Mark
With the rapid development of intelligent driving technology, there has been a growing interest in the driving comfort of automated vehicles. As vehicles become more automated, the role of the driver shifts from actively engaging in driving tasks to that of a passenger. Consequently, the study of the passenger experience in automated driving vehicles has emerged as a significant research area. In order to examine the impact of automatic driving on passengers' riding experience in vehicle platooning scenarios, this study conducted real vehicle experiments involving six participants. The study assessed the subjective perception scores, eye movement, and electrocardiogram (ECG) signals of passengers seated in the front passenger seat under various vehicle speeds, distances, and driving modes. The results of the statistical analysis indicate that vehicle speed has the most substantial influence on passenger perception. The driving mode has a minor effect on the passenger riding experience
Hu, HongyuZhang, GuojuanCheng, MingLi, ZhengyiHe, LeiSu, Lili
Collisions resulting in injuries or fatalities occur more frequently at intersections. This is partly because safe navigation of intersections requires drivers to accurately observe and respond to other road users with conflicting paths. Previous studies have raised questions about how traffic control devices and the positioning of other road users might affect drivers' visual search strategies when navigating intersections. To address these questions, four left-turn-across-path (LTAP) scenarios were created by combining two types of traffic control devices (stop signs and traffic lights) with two hazard starting locations (central and peripheral). Seventy-four licensed drivers responded to all scenarios in a counterbalanced order using a full vehicle driving simulator. Eye-tracking glasses were used to monitor eye movements, both before and after hazard onset. The results revealed that drivers at the signalized intersections took longer to fixate the LTAP hazard before onset, spent
Caren, BrooklinZiraldo, ErikaOliver, Michele
This research investigates the energy savings achieved through eco-driving controls in connected and automated vehicles (CAVs), with a specific focus on the influence of powertrain characteristics. Eco-driving strategies have emerged as a promising approach to enhance efficiency and reduce environmental impact in CAVs. However, uncertainty remains about how the optimal strategy developed for a specific CAV applies to CAVs with different powertrain technologies, particularly concerning energy aspects. To address this gap, on-track demonstrations were conducted using a Chrysler Pacifica CAV equipped with an internal combustion engine (ICE), advanced sensors, and vehicle-to-infrastructure (V2I) communication systems, compared with another CAV, a previously studied Chevrolet Bolt electric vehicle (EV) equipped with an electric motor and battery. The implemented control is a universal speed planner that solves the eco-driving optimal-control problem within a receding-horizon framework
Jeong, JongryeolKandaswamy, ElangovanDudekula, Ahammad BashaHan, JihunKarbowski, DominikNaber, Jeffrey
Driver steering feature clustering aims to understand driver behavior and the decision-making process through the analysis of driver steering data. It seeks to comprehend various steering characteristics exhibited by drivers, providing valuable insights into road safety, driver assistance systems, and traffic management. The primary objective of this study is to thoroughly explore the practical applications of various clustering algorithms in processing driver steering data and to compare their performance and applicability. In this paper, principal component analysis was employed to reduce the dimension of the selected steering feature parameters. Subsequently, K-means, fuzzy C-means, the density-based spatial clustering algorithm, and other algorithms were used for clustering analysis, and finally, the Calinski-Harabasz index was employed to evaluate the clustering results. Furthermore, the driver steering features were categorized into lateral and longitudinal categories. Different
Chen, ChenZong, Changfu
With the revolutionary advancements in modern transportation, offering advanced connectivity, automation, and data-driven decision-making has put the intelligent transportation systems (ITS) to a high risk from being exposed to cyber threats. Development of modern transportation infrastructure, connected vehicle technology and its dependency over the cloud with an aim to enhance safety, efficiency, reliability and sustainability of ITS comes with a lot more opportunities to protect the system from black hats. This paper explores the landscape of cyber threats targeting ITS, focusing on their potential impacts, vulnerabilities, and mitigation strategies. The cyber-attacks in ITS are not just limited to Unauthorized Access, Malware and Ransomware Attacks, Data Breaches, Denial of Service but also to Physical Infrastructure Attacks. These attacks may result in potentially disrupting critical transportation infrastructure, compromise user safety, and can cause economic losses effecting the
Dewangan, Kheelesh KumarPanda, VibekOjha, SunilShahapure, AnjaliJahagirdar, Shweta Rajesh
India is a highly populous country. The traffic problems faced by the people here are not uncommon. The increase in traffic leads to increase in accidents, pollution, inconvenience and frustration. It also comes with costs of additional fuel and time. Though public transport is extensively available in India, still it isn't sufficient for the population of India. Especially in Metro cities, public transport services are often crowded. So, to travel peacefully people are opting for commuting in their own vehicles. And as a result, more vehicles are coming on roads. Other major reasons for increasing traffic are lack of proper implementation of traffic rules and traffic signals out of sync. In addition to city traffic, congestion is also seen on highways, mainly at toll plazas. Although implementation of FASTag has reduced it to some extent, some toll plazas still face traffic congestion issues. This paper provides an idea to ease the traffic problems in the city and on the highways too
Jain, Pritesh
Autonomous cars (ACs) and advanced driver-assistance systems (ADAS) have relied on convolutional neural networks (CNNs) for object detection. However, image degradation caused by adverse weather conditions like rain, snow, and fog can decrease the performance of a CNN. So, this paper presents the development of an image-processing technique aimed to mitigate such a problem. First, after an extensive evaluation of models for object detection, YOLOv3 was chosen because of its compromise between precision and inference time. Afterwards, the training and test of a YOLOv3 CNN was investigated for cars, traffic signals, traffic lights, pedestrians, and riders. Performance was evaluated by estimating the average and mean average precision (mAP) for every one of the mentioned object classes. An OpenCV based pre-processing technique to mitigate the degradation imposed by adverse weather conditions was implemented. Specifically, the OpenCV filters of erosion, dilation and joint bilateral filter
Romão, BrunoFagotto, Eric
The main objective of platoon control is coordinated motion of autonomous vehicle platooning with small intervehicle spacing while maintaining the same speed and acceleration as the leading vehicle, which can save energy consumption and improve traffic throughput. The conventional platoon control methods are confronted with the problem of manual parameter tuning. In order to addres this isue, a novel bifold platoon control approach leveraging a deep reinforcement learning-based model is proposed, which enables the platoon adapt to the complex traffic environment, and guarantees the safety of platoon. The upper layer controller based on the TD3 tuned PID algorithm outputs the desired acceleration. This integration mitigates the inconvenience of frequent manual parameter tuning asociated with the conventional PID algorithm. The lower layer controller tracks the desired acceleration based on the inverse vehicle dynamics model and feedback control. Through this dynamic inverse model, the
Chen, XinhaiWang, RukangCui, YananJin, XiaoxinFeng, ChengjunXie, BoDeng, ZejianChu, Duanfeng
With the extension of intelligent vehicles from individual intelligence to group intelligence, intelligent vehicle platoons on intercity highways are important for saving transportation costs, improving transportation efficiency and road utilization, ensuring traffic safety, and utilizing local traffic intelligence [1]. However, there are several problems associated with vehicle platoons including complicated vehicle driving conditions in or between platoon columns, a high degree of mutual influence, dynamic optimization of the platoon, and difficulty in the cooperative control of lane change. Aiming at the dual-column intelligent vehicle platoon control (where “dual-column” refers to the vehicle platoon driving mode formed by multiple vehicles traveling in parallel on two adjacent lanes), a multi-agent model as well as a cooperative control method for lane change based on null space behavior (NSB) for unmanned platoon vehicles are established in this paper. Specifically, a multi-agent
Yan, DanshuZhao, ZhiguoLiang, KaichongYu, Qin
The development of autonomous driving generally requires enormous annotated data as training input. The availability and quality of annotated data have been major restrictions in industry. Data synthesis techniques are then being developed to generate annotated data. This paper proposes a 2D data synthesis pipeline using original background images and target templates to synthesize labeled data for model training in autonomous driving. The main steps include: acquiring templates from template libraries or alternative approaches, augmenting the obtained templates with diverse techniques, determining the positioning of templates in images, fusing templates with background images to synthesize data, and finally employing the synthetic data for subsequent detection and segmentation tasks. Specially, this paper synthesizes traffic data such as traffic signs, traffic lights, and ground arrow markings in 2D scenes based on the pipeline. The effectiveness of this pipeline was verified on the
Bie, XiaofangZhang, SongMeng, ChaoMei, JinrenLi, JianHe, Xin
Vehicle-to-infrastructure (V2I) connectivity technology presents the opportunity for vehicles to perform autonomous longitudinal control to navigate safely and efficiently through sequences of V2I-enabled intersections, known as connected corridors. Existing research has proposed several control systems to navigate these corridors while minimizing energy consumption and travel time. This article analyzes and compares the simulated performance of three different autonomous navigation systems in connected corridors: a V2I-informed constant acceleration kinematic controller (V2I-K), a V2I-informed model predictive controller (V2I-MPC), and a V2I-informed reinforcement learning (V2I-RL) agent. A rules-based controller that does not use V2I information is implemented to simulate a human driver and is used as a baseline. The performance metrics analyzed are net energy consumption, travel time, and root-mean-square (RMS) acceleration. Two connected corridor scenarios are created to evaluate
King, BrianOlson, JordanHamilton, KaylaFitzpatrick, BenjaminYoon, Hwan-SikPuzinauskas, Paul
The cooperative platoon of multiple trucks with definite proximity has the potential to enhance traffic safety, improve roadway capacity, and reduce fuel consumption of the platoon. To investigate the truck platooning performance in a real-world environment, two Peterbilt class-8 trucks equipped with cooperative truck platooning systems (CTPS) were deployed to conduct the first-of-its-kind on-road commercial trial in Canada. A total of 41 CTPS trips were carried out on Alberta Highway 2 between Calgary and Edmonton during the winter season in 2022, 25 of which were platooning trips with 3 to 5 sec time gaps. The platooning trips were performed at ambient temperatures from −24 to 8°C, and the total truck weights ranged from 16 to 39 tons. The experimental results show that the average time gap error was 0.8 sec for all the platooning trips, and the trips with the commanded time gap of 5 sec generally had the highest variations. The average number of disengagements increased when the
Jiang, LuoKheyrollahi, JavadKoch, Charles RobertShahbakhti, Mahdi
Items per page:
1 – 50 of 498