Dense Traffic Scenarios Research Articles

Behavior recognition of non-motorized transport at intersections using dual-channel grid model based on disordered trajectory point data

Accurately identifying the passing and waiting behavior of pedestrians and non-motorized vehicles at intersection is essential to planning management measures for non-motorized transport. Much previous study in this field has focused on methods based on continuous individual trajectory detection, which are almost ineffective under the poor detection condition in dense traffic scenarios. To address the task, this paper establish a workable framework for inferring the probabilities of passing or waiting behavior of pedestrians and non-motorized vehicles in arbitrary space using disordered trajectory point data. First, a two-channel model is proposed to perform a formalized grid-level representation of an intersection, in which the functional attributes and occupancy characteristics of grids are comprehensively defined and quantified. Then, the quantitative characteristics of the grids are used to detect the real-world occurrence space of passing and waiting behaviors, by the clustering and expanding operations on grids. Finally, through feature transfer along path, characteristics is decomposed into passing and waiting occurrence characteristics for behavior probability computation. Results indicate that the method achieves over 91% accuracy of behavior recognition, which is better than compared methods in various Multiple Object Tracking Accuracy (MOTA). Although the method is sensitive to spatial detection conditions, it obtains steady accuracy under various target detection settings.

Merging planning in dense traffic scenarios using interactive safe reinforcement learning

Autonomous navigation in dense traffic scenarios, such as on-ramp forced merging, still poses significant challenges for autonomous vehicles to prevent accidents and alleviate traffic congestion. This paper introduces a novel motion planning framework that combines Interactive Safe Reinforcement Learning (IntSRL) with Nonlinear Model Predictive Control (NMPC). This framework develops an interactive merging planning policy that accounts for the uncertainty of traffic participants, multi-objective optimization and heterogeneous vehicle interactions, in which the upper planner, i.e., IntSRL, furnishes the lower planner, NMPC, with global guidance path and velocity guidance. An Adaptive Safety Governor (ASG) module within IntSRL adjusts potentially unsafe actions by incorporating prior knowledge and driving experience. And a coupling evaluation mechanism for multi-objective optimization is embedded into reward shaping with integration of driving safety and strategy efficiency. We evaluate the proposed controller on various dense traffic scenarios using the proposed Heterogeneous Intelligent Driver Model (H-IDM) considering different driving styles and cooperative willingness of other vehicles. The test results indicate that the proposed method surpasses existing optimization-based and learning-based baselines in qualitative and quantitative measures.

Augmented driver behavior models for high‐fidelity simulation study of crash detection algorithms

AbstractDeveloping safety and efficiency applications for Connected and Automated Vehicles (CAVs) requires a great deal of testing and evaluation. The need for the operation of these systems in critical and dangerous situations makes the burden of their evaluation very costly, possibly dangerous, and time‐consuming. As an alternative, researchers attempt to study and evaluate their algorithms and designs using simulation platforms. Modeling the behavior of drivers or human operators in CAVs or other vehicles interacting with them is one of the main challenges of such simulations. While developing a perfect model for human behavior is a challenging task and an open problem, a significant augmentation of the current models used in simulators for driver behavior is presented. In this paper, a simulation framework for a hybrid transportation system is presented that includes both human‐driven and automated vehicles. In addition, the human driving task is decomposed and a modular approach is offered to simulate a large‐scale traffic scenario, allowing for a thorough investigation of automated and active safety systems. Such representation through Interconnected modules offers a human‐interpretable system that can be tuned to represent different classes of drivers. Additionally, a large driving dataset is analyzed to extract expressive parameters that would best describe different driving characteristics. Finally, a similarly dense traffic scenario is recreated within the simulator and a thorough analysis of various human‐specific and system‐specific factors is conducted, studying their effect on traffic network performance and safety.

Dual-Horizon Reciprocal Collision Avoidance for Aircraft and Unmanned Aerial Systems

The aircraft conflict detection and resolution problem has been addressed with a wide range of centralised methods in the past few decades, e.g. constraint programming, mathematical programming or metaheuristics. In the context of autonomous, decentralized collision avoidance without explicit coordination, geometric methods provide an elegant, cost-effective approach to avoid collisions between mobile agents, provided they all share a same logic and a same view of the traffic. The Optimal Reciprocal Collision Avoidance (ORCA) algorithm is a state-of-the art geometric method for robot collision avoidance, which can be used as a Detect & Avoid logic on-board aircraft or Unmanned Aerial Vehicles. However, ORCA does not handle well some degenerate situations where agents operate at constant or near-constant speeds, which is a widespread feature of commercial aircraft or fixed-winged Unmanned Airborne Systems. In such degenerate situations, pairs of aircraft could end up flying parallel tracks without ever crossing paths to reach their respective destination. The Constant Speed ORCA (CS-ORCA) was proposed in 2018 to better handle these situations. In this paper, we discuss the limitations of both ORCA and CS-ORCA, and introduce the Dual-Horizon ORCA (DH-ORCA) algorithm, where two time horizons are used respectively for short-term collision avoidance and medium-term path-crossing. We show that this new approach mitigates the main issues of ORCA and CS-ORCA and yields better performances with dense traffic scenarios.

Learning Interaction-Aware Guidance for Trajectory Optimization in Dense Traffic Scenarios

Autonomous navigation in dense traffic scenarios remains challenging for autonomous vehicles (AVs) because the intentions of other drivers are not directly observable and AVs have to deal with a wide range of driving behaviors. To maneuver through dense traffic, AVs must be able to reason how their actions affect others (interaction model) and exploit this reasoning to navigate through dense traffic safely. This paper presents a novel framework for interaction-aware motion planning in dense traffic scenarios. We explore the connection between human driving behavior and their velocity changes when interacting. Hence, we propose to learn, via deep Reinforcement Learning (RL), an interaction-aware policy providing global guidance about the cooperativeness of other vehicles to an optimization-based planner ensuring safety and kinematic feasibility through constraint satisfaction. The learned policy can reason and guide the local optimization-based planner with interactive behavior to pro-actively merge in dense traffic while remaining safe in case other vehicles do not yield. We present qualitative and quantitative results in highly interactive simulation environments (highway merging and unprotected left turns) against two baseline approaches, a learning-based and an optimization-based method. The presented results show that our method significantly reduces the number of collisions and increases the success rate with respect to both learning-based and optimization-based baselines.

Intention-Aware Interactive Transformer for Real-Time Vehicle Trajectory Prediction in Dense Traffic

The intelligent transportation system (ITS) is one of the key components of future transportation. Vehicle trajectory prediction, which is made possible by recent advancement in sensor and communication technology, can help ITS to improve efficiency and safety of transportation by detecting potential conflicts and accidents. However, prior works mainly focus on trajectory prediction for autonomous vehicles in small-scale scenarios. In addition, interpretability of trajectory prediction, which facilitates trustworthiness of ITS, has received scant attention in the research literature. This study proposes intention-aware interactive transformer (IIT) model to address the problem of real-time vehicle trajectory prediction in large-scale dense traffic scenarios. IIT follows the transformer architecture to process time-series data and adopts an intention-based scheme to predict future trajectories. Unlike prior works that use complicated data structure to represent the spatial relation between vehicles, what is unique about IIT is that the interactions between vehicles are captured using a multi-head attention (MHA) mechanism. Moreover, the interpretability of IIT is illustrated in two levels: inter-vehicle attention and intra-vehicle intention. Experimental results suggest that MHA significantly reduces the computational cost of data preprocessing for IIT. Consequently, IIT runs up to eight times faster than baseline models in large-scale dense traffic scenarios while only suffering an average of 7.6% accuracy degradation in a 5-s prediction horizon. Further, interpretability analysis is conducted, which reveals that the interpretability of IIT is beneficial to ITS in many ways, such as detecting potential congestion and solving driver conflicts.

TRAP: Traffic-Based Adaptive Ramp Packing for Blind Cancellation in Autonomous Vehicles

Autonomous Vehicles (AVs) rely on a set of radar sensors used to map surrounding environment. Most commonly used radar sensors for AVs use Frequency Modulated Continuous Wave (FMCW) ramps for object parameter (i.e., range and relative velocity) estimation. Due to large bandwidth requirement of FMCW radar, only a limited number of AVs can be operated in the available spectrum. Consequently, the co-existence of large number of AVs may lead to the problem of radar-to-radar interference, also referred to as radar blindness. Moreover, the problem becomes more severe in the higher traffic scenarios. In this work, we propose a Traffic-based Adaptive Ramp Packing (TRAP) scheme, which adapts radar range and assigns FMCW ramp parameters on the basis of inter-vehicular distance among AVs. Specifically, TRAP scheme allows to make effective use of the available time-frequency resource, and enables to pack more ramps in the dense traffic scenarios. Further, it is shown that adaptive radar range adoption may provide significantly more number of ramps in the given bandwidth. Furthermore, through simulation results, it is shown that TRAP significantly reduces the blind probability against state-of-the-art fixed range schemes.

Spatiotemporal Costmap Inference for MPC Via Deep Inverse Reinforcement Learning

It can be difficult to autonomously produce driver behavior so that it appears natural to other traffic participants. Through Inverse Reinforcement Learning (IRL), we can automate this process by learning the underlying reward function from human demonstrations. We propose a new IRL algorithm that learns a goal-conditioned spatio-temporal reward function. The resulting costmap is used by Model Predictive Controllers (MPCs) to perform a task without any hand-designing or hand-tuning of the cost function. We evaluate our proposed Goal-conditioned SpatioTemporal Zeroing Maximum Entropy Deep IRL (GSTZ)-MEDIRL framework together with MPC in the CARLA simulator for autonomous driving, lane keeping, and lane changing tasks in a challenging dense traffic highway scenario. Our proposed methods show higher success rates compared to other baseline methods including behavior cloning, state-of-the-art RL policies, and MPC with a learning-based behavior prediction model.

R-Comm: A Traffic Based Approach for Joint Vehicular Radar-Communication

Automotive radar and vehicular communication are the two primary means of establishing an intelligent transportation system. However, both the systems are susceptible to interference. The inter-vehicular interference significantly affects the radar and communication performance, especially in the dense traffic scenarios. In this work, we have shown that lowering down radar range in dense traffic scenario provides twofold advantages; (a) it reduces radar-to-radar interference and, (b) it provides resources to support the vehicular communication. We propose a joint Radar-Communication (R-Comm) algorithm which enables connected vehicles to use a fraction of radar resources for vehicular communication based on the traffic density. Also, two different schemes have been proposed for R-Comm transmission in the sparse and dense traffic scenarios. Further, through simulation results, it is shown that R-Comm benefits both radar and communication systems.

Unreliable V2X Communication in Cooperative Driving: Safety Times for Emergency Braking

Cooperative driving is a promising paradigm to improve traffic efficiency and safety. In congested traffic scenarios, such cooperation allows for safe maneuvering and driving with small inter-vehicle spatial gaps. The vehicles involved coordinate their movements in real-time and continuously update each other about their maneuver execution status by means of Vehicle-to-Everything (V2X) communication. However, unreliable V2X communication increases the Age of Information (AoI) of vehicles’ status updates, posing a challenge in situations where emergency braking is required during cooperative maneuvering. To address the interplay between unreliable V2X communication and the resulting impact on traffic safety, we introduce a so-called safety time function , specifically designed for cooperative driving use-cases. The safety time function provides the time available for a vehicle to react to an unexpected event of another vehicle – such as emergency braking to avoid a collision. We provide a computationally efficient algorithm for the computation of safety time functions, which allows for efficient and safe cooperative maneuver planning – even in dense traffic scenarios with many vehicles involved. We show the applicability of our proposed safety time function based on the assessed communication quality for IEEE 802.11p-based V2X communication to meet safety constraints in dense vehicular traffic.

A Fuzzy-Based Congestion Control Scheme for Vehicular Adhoc Network Communication

Vehicular ad hoc network (VANET) is a self-organized, multi-purpose, service-oriented communication network that enables communication between vehicles and between vehicles and roadside infrastructures for the purpose of exchanging messages. In a dense traffic scenario, the message traffic may generate a load higher than the available capacity of the transmission medium leading to channel congestion problem. This situation leads to a rise in packet loss rates and transmission delay. Some existing congestion control schemes adapt the transmission power, transmission rate, and contention window parameters by making comparison with neighboring values through classical logic. However, the approach does not consider points between two close parameter values. This work uses fuzzy logic to improve the adaptation process of the network contention window parameter. The proposed scheme achieved a 15% higher in-packet delivery ratio and 10ms faster transmission compared with related work in terms end-to-end delay.

Bi-Directional Dense Traffic Counting Based on Spatio-Temporal Counting Feature and Counting-LSTM Network

Machine vision based vehicle counting and traffic flow estimation are challenging problems especially for dense traffic scenarios. Previous <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">line of interest</i> (LOI) counting methods rarely focus on dense scenarios and their performance largely relies on the accuracy of tracking. Avoiding the use of complex tracking methods, an LOI counting framework is proposed to address the bi-directional LOI counting problem in dense scenarios. There are three main contributions. Firstly, instead of treating the LOI vehicle counting problem as a combination of detecting and tracking of individual vehicles, the bi-directional traffic flow is taken as a whole and a novel <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">spatio-temporal counting feature</i> (STCF) is proposed for extracting bi-directional traffic flow features in dense traffic scenarios. Secondly, without relying on a multi-target tracking process for tracking and counting each vehicle, a counting network is proposed, called the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">counting Long Short-Term Memory</i> (cLSTM) network, to do analysis of the bi-directional STCF features and vehicle counting in successive video frames. Lastly, an estimation model is designed for estimating traffic flow parameters including speed, volume and density. Experiments performed on the UA-DETRAC dataset and the captured videos show that the proposed vehicle counting method outperforms the tested representative LOI counting methods in both accuracy and speed, and that the proposed framework can efficiently estimate traffic flow parameters including speed, volume and density in real time.

WiFi-Assisted Control Traffic Offloading in Vehicular Environments

Targeting mobile vehicle communications, Dedicated Short Range Communications (DSRC) wireless technology has been widely used in Vehicular Ad-hoc Networks (VANETs). Though DSRC can provide reliable and low-latency communications for various VANET applications, the multi-channel operations standardized by IEEE 802.11p suffer from high contention on control channel, especially in dense traffic scenarios. Meanwhile, as a conventional wireless technology for VANET communications, WiFi is often underutilized when a vehicle is on road. In this paper, we propose an innovative approach that can effectively offload control frames from DSRC to WiFi when the control traffic is dense, resulting in significantly improved throughput of control traffic (about 28.7% under moderate traffic density in simulation).

Self-Assessment Based Clustering Data Dissemination for Sparse and Dense Traffic Conditions for Internet of Vehicles

Internet of Vehicle (IoV) is a sub class of vehicular ad hoc networks with more advanced cloud and Internet-enabled services. These networks offer various types of safety and infotainment services and provide comfortability and safety to passengers as well as to the drivers. Due to the high mobility of nodes, the nodes are out from its communication range and the information becomes outdated and causes link disconnections and packet dropping. Most feasible routing protocols are needed to provide in-time data communication, handle high mobility of nodes, dynamic topologies and unpredictable environments of these networks. In this paper, we proposed SACBR (Self-Assessment Cluster-based Routing) protocol in which the Cluster Heads (CHs) can communicate with other CHs. Every vehicle node initiates a self-assessment approach based on more appropriate routing metrics and elects the CH for every cluster and then collects the data from member nodes and further forward the data to other CHs. The CH is responsible to manage its own and member nodes’ data forwarding process. The proposed protocol provides more stability and less overhead compared to the aggregation method where every node exchanges its data with a one-hop neighbor. Proposed protocol suites sparse and dense traffic scenarios where most of the time vehicle nodes are moving in platoons or snaking structures. The experimental results show the better performance of SACBR compared to state-of-the-art protocols.

RBO-EM: Reduced Broadcast Overhead Scheme for Emergency Message Dissemination in VANETs

Effective information dissemination constitutes the cornerstone of communication in Vehicular Ad hoc Networks (VANETs). Avoiding broadcast storm is one of the considerable challenges in the development of an effective dissemination scheme for VANETs, which leads to extensive rebroadcasting. Reliable emergency messages delivery is another substantial measure of performance, especially in dense traffic and high mobility scenarios. A generic solution to deal with these problems is to cluster vehicles, so that emergency messages can be re-broadcasted to adjacent clusters only by a smaller number of vehicles. Thus, reliable selections of cluster head and relay nodes are important to reduce the number of retransmissions. In this regard, we propose a clustering technique for Reduced Broadcast Overhead Scheme for Emergency Message Dissemination (RBO-EM). RBO-EM is based on the mobility metrics to strengthen the cluster formation, avoid communication overhead and maintain the message reliability in a high mobility scenario. Moreover, we introduce relay nodes based on estimated link stability to limit the number of retransmissions. RBO-EM is evaluated against eminent schemes by varying traffic densities and speed. Simulation results indicate that RBO-EM enables a reasonable performance gain in terms of coverage, end-to-end delay and message reliability.

Simultaneous Visible Light Communication and Distance Measurement Based on the Automotive Lighting

Visible light communication could be an interesting complement to the IEEE 802.11p-based vehicle-to-vehicle communication systems that are sensitive to interferences and delays in dense traffic scenarios such as platooning. Visible light could also provide a redundant distance measure that is crucial for path control in this application. In this paper, a system called visible light communication rangefinder and performing simultaneously vehicle-to-vehicle communication and range-finding using the headlamps and taillights is proposed for the first time. By exchanging a clock signal contained in Manchester-encoded signals, both the following and leading vehicles can share information and estimate their inter-distance through phase-shift measurement. The system is first presented theoretically and it is shown in particular that the Doppler effect has no significant impact on both functions. Then, it is modeled and validated using Simulink. Finally, both the range-finding and communication function are validated experimentally. The range-finding function is functional up to 25 m, and has a resolution of around 24 cm at 10 m, whereas the communication function provides a 500 kbps link with a BER below $10^{-6}$ up to 30 m.

Test analysis of a scalable UAV conflict management framework

This study elaborates the conflict management framework of unmanned aerial vehicles, focusing on the identification of the spatiotemporal interdependencies between them, with consideration of the future scalability problems in highly dense traffic scenarios. The paper first tries to justify the applied separation criteria among small cooperative unmanned aerial vehicles based on their performance characteristics and the planned missions’ type. The adopted criteria, obtained from the simulations of 160 missions, present a testing asset, referring to a current lack of the spatiotemporal requirements and a need for extending the research in this area to provide a more rapid integration of these vehicles into the civil controlled airspace. The paper then elaborates the computational framework for the conflict detection and resolution function and operational metrics for causal identification of the spatiotemporal interdependencies between two or more cooperative vehicles. The vehicles are considered as a conflict mission system that strives to achieve an efficient solution by applying certain maneuvering measures, before a loss of separation occurs. The operational trials of five local, short-range missions, supported by the simulation scenario, demonstrate the potential for a time-based complexity analysis in the conflict resolution processes with less demanding and more efficient coordinated maneuvers. The results show that those maneuvers would not induce any new conflicts and disrupt the cooperative mission system when the spatial capacity only might not be favorable in provision of the avoidance maneuvers within an available airspace.

Plummeting Broadcast Storm Problem in Highways by Clustering Vehicles Using Dominating Set and Set Cover.

“Vehicular Ad-hoc Networks” (VANETs): As an active research area in the field of wireless sensor networks, they ensure road safety by exchanging alert messages about unexpected events in a decentralized manner. One of the significant challenges in the design of an efficient dissemination protocol for VANETs is the broadcast storm problem, owing to the large number of rebroadcasts. A generic solution to prevent the broadcast storm problem is to cluster the vehicles based on topology, density, distance, speed, or location in such a manner that only a fewer number of vehicles will rebroadcast the alert message to the next group. However, the selection of cluster heads and gateways of the clusters are the key factors that need to be optimized in order to limit the number of rebroadcasts. Hence, to address the aforementioned issues, this paper presents a novel distributed algorithm CDS_SC: Connected Dominating Set and Set Cover for cluster formation that employs a dominating set to choose cluster heads and set covering to select cluster gateways. The CDS_SC is unique among state-of-the-art algorithms, as it relies on local neighborhood information and constructs clusters incrementally. Hence, the proposed method can be implemented in a distributed manner as an event-triggered protocol. Also, the stability of cluster formation is increased along with a reduction in rebroadcasting by allowing a cluster head to be passive when all its cluster members can receive the message from the gateway vehicles. The simulation was carried out in dense, average, and sparse traffic scenarios by varying the number of vehicles injected per second per lane. Besides, the speed of each individual vehicle in each scenario was varied to test the degree of cohesion between vehicles with different speeds. The simulation results confirmed that the proposed algorithm achieved 99% to 100% reachability of alert messages with only 6% to 10% of rebroadcasting vehicles in average and dense traffic scenarios.

An evaluation of factors influencing drag coefficient in double-deck tunnels by CFD simulations using factorial design method

Vehicular drag coefficients in road-tunnel systems are complexly affected by traffic conditions, in terms of vehicle speed and density. Using computational fluid dynamics (CFD)-based factorial design, this study evaluates factors influencing the drag coefficient of a sedan in the Korean double-deck tunnel (KDDT) system, which divides the roadway into upper and lower levels within a single-tube tunnel. The tested factors included blockage ratio, Reynolds number, spacing ratio, and roughness height of the vehicular surface. Results showed that the blockage ratio had the most significant effect in the KDDT system, followed by the spacing ratio and the Reynolds number. A negative interaction effect was observed between the Reynolds number and spacing ratio, thereby indicating that dense traffic scenarios with fast-moving dense vehicles serve to reduce the averaged drag coefficient. In conjunction with the face-centered cube design, a response surface model (RSM) was constructed using a second-order polynomial model. For various tunnel configurations, data predicted by RSM were compared against results obtained from CFD simulations. A good agreement was observed between CFD and RSM with relative errors not exceeding 8.3% in most tunnel configurations investigated. Findings of this study suggest that the RSM can provide accurate estimates of the drag coefficient.

Trust-based security adaptation mechanism for Vehicular Sensor Networks

Vehicular Sensor Networks (VSNs) are foreseen as a promising technology that can provide safe and reliable road travel in smart cities. By establishing pervasive connectivity among vehicles and infrastructure units on the road, various intelligent transport applications could be realized. Efficient network security for VSNs thus becomes a key challenge to reliably implement these applications. However, robust security techniques incur high security overhead and processing delays which significantly impact Quality of Service (QoS) particularly in a dense traffic scenario. In this paper, we propose a trust-based security adaptation mechanism to improve the QoS of safety applications in VSNs. The trust level is calculated using connectivity duration, security level and centrality metrics of nearby vehicles. The simulation results we have obtained with our proposed research shows an improvement of 25–65% in terms of safety awareness and 33–53% improvement in terms of packet inter-arrival time of safety applications in VSNs.