Vehicle Trajectory Research Articles

A Comprehensive Investigation of Lane-Changing Risk Recognition Framework of Multi-Vehicle Type Considering Key Features Based on Vehicles’ Trajectory Data

To comprehensively investigate the key features of lane-changing (LC) risk for different vehicle types during left and right LC, and to improve the accuracy of LC risk recognition, this paper proposes a key feature selection and risk recognition model based on vehicle trajectory data. Based on a HighD high-precision vehicle trajectory dataset, the trajectory data of LC vehicles and surrounding vehicles of each vehicle type are extracted. SDI (stop distance index) and CI (crash index) are selected as surrogate indicators to calculate the risk exposure level (REL) and risk severity level (RSL). The K-means algorithm is used to cluster the REL and RSL to obtain the LC risk level, which is divided into three levels. The combination of basic features and interaction features of LC vehicles and surrounding vehicles with LC risk levels is constructed as the LC risk feature dataset. Based on the LightGBM (light gradient boosting machine) algorithm, the importance of features is sorted. Finally, a CNN-BiLSTM-Attention model is established to recognize the LC risk of each vehicle type during left and right LC. The results indicate that significant differences exist among different vehicle types and LC directions. Compared with CNNs (convolutional neural networks), LSTM (long short-term memory), and BiLSTM (bi-directional long short-term memory), CNN-BiLSTM-Attention performs best in recognizing the risk of LC in all cases. Moreover, the key feature groups that have the optimal result of recognizing the risk of LC in different cases are obtained.

Cooperative control strategy of trajectory tracking and driving stability for distributed-drive vehicles under extreme conditions

Active chassis dynamics control has become increasingly important in recent years for achieving high-level automated driving of intelligent electric vehicles. To improve tracking performance of the commonly-used model predictive control (MPC) in extreme scenarios, such as emergency lane changing, an improved MPC motion stability controller (MPC_ISMC) based on sliding mode variable structure active steering control and differential torque yaw control is proposed. Firstly, an MPC-based trajectory tracking controller is designed to track the desired trajectory. Subsequently, active front-wheel steering (AFS) and direct yaw moment (DYC) controllers are designed by using the backstepping sliding mode variable structure and the adaptive super-twisting sliding mode variable structure, respectively. Moreover, a cooperative control law for the AFS and the DYC is formulated by the smooth switching function and intervention intensity factor with respect to vehicle stable state and trajectory tracking error. After that, a torque distribution scheme of four wheels is optimized to achieve the additional yaw moment target. Finally, the effectiveness of the proposed cooperative control strategy is validated using a hardware-in-the-loop platform. The results indicate that the proposed controller has better performance in trajectory tracking and vehicle stability under extreme conditions compared with traditional MPC and other methods.

An effective lane changing behaviour prediction model using optimized CNN and game theory

Accurately predicting lane changes, a crucial driving activity for preventing accidents and ensuring driver safety, is addressed in this study. An innovative predictive model that integrates game theory for precise lane change intention detection and an optimized Convolutional Neural Network (CNN) for trajectory prediction is proposed in this study. The CNN's efficiency is enhanced through metaheuristic optimization of both the convolution and fully connected layers using the Whale Optimization Algorithm (WOA). Emphasizing robust data processing, a Wiener filter is applied for pre-processing, and the Cascaded Fuzzy C means (CFCM) technique is employed for segmentation. The resulting Whale Optimization Algorithm-based CNN (WOA-CNN) effectively forecasts the trajectory of lane-changing vehicles. Validation of the proposed approach in Python demonstrates exceptional accuracy, reaching 96.5%. This study showcases the effectiveness of the WOA-CNN model in advancing the prediction accuracy of lane-changing behaviour, contributing to enhanced driver safety and accident prevention.

Optimization of isolated intersection signal timing and trajectory planning under mixed traffic environment: The flexible catalysis of connected and automated vehicles

With the development of connected and automated vehicles (CAVs) enables real-time interaction between intersection signals and vehicle trajectories. The full use of this technological breakthrough will help traffic managers improve the performance of intersections. Most previous research focuses on the 100% CAV environment and the single lateral or longitudinal optimization of CAVs. Based on the flexible characteristics of CAVs, this paper proposes that CAV is regarded as the catalyst of vehicle platoon in mixed traffic environment and considers the uncertainty of CAVs and CHVs’ interaction in the real-world, which can promote the generation of controllable platoon through the “catalytic” mode of cooperative, accelerated, or direct lane-changing, and cooperative or direct overtaking. Simultaneously, a two-stage optimization model of mixed traffic trajectory and signal timing is proposed. Stage I: Based on the predicted vehicle platoon information, a dynamic NEMA signal timing scheme without duplicate structures is generated to minimize vehicle delays. Stage II: Based on the timing scheme, the generation of controllable and stable platoon and vehicle trajectory optimization model are established to minimize vehicle emissions. Dynamic programming with NEMA signal groups as sub-states is designed to solve the proposed model. The performance of the proposed model under different scenarios is investigated through numerical experiments and compared with benchmark models. Results show that the proposed model will outperform the benchmark models regarding average vehicle delay and emission under more realistic traffic demands. The average vehicle delay can be reduced by 54.32% and 7.33%, and the average vehicle emissions can be reduced by 19.1% and 0.8%, respectively. Meanwhile, the sensitivity analysis of CAV market penetration shows that the proposed model can perform satisfactorily at 20% CAV market penetration. Notedly, with increased market penetration, the proposed model will obtain better performance.

A microscopic traffic flow model for explaining nonlinear traffic phenomena: Modeling, stability analysis and validation

As the core of reproducing the real nonlinear phenomena of traffic congestion, the investigation of car-following models, which take into account human factors, has consistently assumed a central role within the domain of transportation science. Nevertheless, it is noteworthy that prior studies have invariably employed a fixed value for driver reaction time to stimuli, whereas it is imperative to recognize the dynamic nature of this parameter, closely associated with the instantaneous vehicle speed. To address this issue, this paper constructs a dynamic sensitivity coefficient (DSC), and further develops a nonlinear human factoring car-following model to reveal the relation between the reaction time and the current speed. Firstly, the linear critical stability condition and the mKdV equation of the model are derived by the linear stability analysis and kink–antikink solution analytic method, respectively. Then, the numerical experiments are conducted to demonstrate that the proposed model is more practical and can reproduce the real driving behavior and phenomena. Finally, calibration and validation results exhibit the actual vehicle trajectory can be simulated well.

Predicting vehicle trajectory of non-lane based driving behaviour with Temporal Fusion Transformer

Complex road sections without lane markings cannot constrain vehicles to follow the lane disciplines. As a result, vehicles often exhibit more disorderly rapid lateral movements (RLMs) in these areas, making it difficult to accurately predict vehicle trajectories. This study takes toll plaza diverging area as an example to propose a framework incorporated the Hidden Markov Model (HMM) and Temporal Fusion Transformer (TFT) for vehicle trajectory prediction in non-lane based complex road sections. The results demonstrate that the vehicles exhibit more RLMs when there are more toll lanes matching their toll collection types. Validated on two toll plaza diverging areas with different structures, the proposed framework achieves higher prediction accuracy than other state-of-the-art predictive methods, particularly in long prediction horizons. In addition, the interpretability of TFT suggests that incorporating RLM intention prediction and environmental factors specific to non-lane based areas into trajectory prediction is of great importance.

From yachts to spacecraft: Legal aspects of space tourism ventures in comparison with the legal regime governing marine tourism

Until recently space travel was considered a privilege of governmental astronauts, with the exception perhaps of a few billionaire space tourists, who had been accommodated on the International Space Station on some rare occasions. Nonetheless, with major space pioneers announcing plans for space tourism activities and after the successful launch in orbit of space tourism vehicles from the Virgin Galactic and Blue Origin companies, academic interest about human spaceflight for purposes of leisure and recreation has ignited. Taking into account that space tourism is in fact an unregulated activity, this paper will firstly examine whether space tourism activities fall under the scope of space law or under the scope of international air law, depending on the planned trajectory of the space tourism vehicle. Furthermore, the authors will attempt to draw conclusions concerning the legal implications of space tourism activities through a comparative analysis with the 1974 Athens Convention on the Carriage of Passengers and the Carriage of their Luggage by Sea. This international convention provides a great example on the regulation of touristic ventures, as it refers to commercial touristic activities realized within a res communis omnium, i.e., the high seas, similarly with touristic activities in space. Moreover, said Convention provides for a specialized regime for touristic activities and is of particular interest, since it has achieved stability and a desirable balance of interests between private tourism operators and passengers. In this connection, the authors will evaluate the provisions of the 1974 Athens Convention in comparison with the space treaties applicable to space tourism activities. Furthermore, suggestions will be made as to whether it would be appropriate and feasible to introduce in the space law realm certain rules of law similar to those of the 1974 Athens Convention, in order to facilitate growth in the space tourism sector, without compromising the safety of human spaceflight for space passengers.

Advancing the white phase mobile traffic control paradigm to consider pedestrians

AbstractCurrent literature on joint optimization of intersection signal timing and connected automated vehicle (CAV) trajectory mostly focuses on vehicular movements paying no or little attention to pedestrians. This paper presents a methodology to safely incorporate pedestrians into signalized intersections with CAVs and connected human‐driven vehicles (CHVs). The movements of vehicles are controlled using both traffic lights and mobile CAV controllers during our newly introduced “white phase.” CAVs navigate platoons of CHVs through the intersection when the white phases are active. In addition to optimizing CAV trajectories, the model optimally selects the status of the traffic light signal among white and green indications for vehicular and walk and do‐not‐walk intervals for pedestrian movements. A receding horizon‐based methodology is used to capture the stochastic nature of the problem and to reduce computational complexity. The case study results show the successful operation of fleets consisting of pedestrians, CAVs, and CHVs with various demand levels through isolated intersections. The results also show that increasing the CAV market penetration rate (MPR) can decrease average intersection delay by up to 27%. Moreover, the average pedestrian, CHV, and CAV delays decrease as the CAV MPR increases and reach their minimum values with a fully CAV fleet. In addition, the presence of the white phase can decrease the intersection average delay by up to 14.7%.

Constrained multi-objective trans-media maneuver trajectory optimization for morphing unmanned aerial-underwater vehicles

In this study, the challenge of optimizing trans-media maneuver trajectories is addressed through the introduction of a Constrained Multi-Objective Grey Wolf Optimizer (CMOGWO). The objective function has been refined to assess solutions more accurately based on their feasibility and adherence to constraints. Three distinct Archives have been employed for categorizing and optimizing dominant, non-dominant, and feasible non-dominant solutions, each benefiting from specifically tailored optimization strategies. A hybrid search method, complemented by a watch-based strategy, is proposed for effectively guiding the population towards the Pareto front, ensuring the discovery of potential optimal solutions. Investigations have been conducted into the impact of initial altitude on trajectory optimization, with comparative simulations revealing the effectiveness of the proposed CMOGWO in managing unmanned aerial-underwater vehicle trajectories. The results demonstrate the potential of the CMOGWO in enhancing the reliability and feasibility of trajectory optimization in complex environmental conditions.

Investigation of Surface-Catalycity Effects on Hypersonic Glide Vehicle Trajectory Optimization

The optimization of a hypersonic glide vehicle’s reachable domain is studied using different models to constrain the stagnation-point heat transfer. The vehicle is modeled as a high-lift common aero vehicle that uses bank-angle modulation for crossrange maneuvering with a leading-edge thermal protection system made of an ultra-high-temperature ceramic. A constrained nonlinear optimization problem is formulated to determine the range of feasible trajectories that satisfy aerodynamic heating and loading constraints. The optimal control profile is determined using a modified particle swarm optimization routine that employs a penalty function approach to handle path and terminal constraints. Baseline trajectories are determined using an existing convective heat-flux correlation, and the effect of surface catalycity on the optimized trajectories is investigated by applying a correction factor, derived from high-fidelity computational fluid dynamics simulations, to the heat-flux correlation. Results demonstrate a high sensitivity of the optimized trajectories to the underlying aerothermodynamics model such that accounting for surface catalycity expands the reachability by an order of magnitude.

Hierarchical vector transformer vehicle trajectories prediction with diffusion convolutional neural networks

In dynamic and interactive autonomous driving scenarios, accurately predicting the future movements of vehicle agents is crucial. However, current methods often fail to capture trajectory uncertainty, leading to limitations in trajectory prediction performance. To address these limitations, this paper introduces the hierarchical vector transformer diffusion model, a novel trajectory prediction method that prioritizes both speed and accuracy. The proposed model decomposes the traffic scene modeling into local patches and global interactions, allowing for the acquisition of relevant environmental and global information. Moreover, a local diffusion encoder is employed to effectively capture the aleatoric uncertainty. The proposed model utilizes an adaptive graph structure to exploit the spatial and temporal relationships inherent in the trajectory data. By employing a graph diffusion process, the model effectively captures dynamic features from the historical trajectory information. Moreover, the model demonstrates adaptability by dynamically adjusting to diverse trajectory data and scenarios, thereby enabling the generation of predicted trajectories that are uncertainty aware. This approach contributes to more effective and efficient modeling of dynamic autonomous driving scenarios. Experimental results demonstrate the superior speed and accuracy of the proposed method compared to existing approaches for trajectory prediction. The proposed method significantly enhances prediction accuracy, achieving results of ADE 0.68 and FDE 1.02 on the Argoverse dataset. In comparison to the baseline model, there are notable improvements in ADE and FDE by 0.03 and 0.06, respectively. It is noteworthy that this method also reduces the inference time by 7% when compared to the currently fastest method.

A communication link lifetime prediction-supported V2V partial computation offloading scheme for autonomous driving

Vehicle Edge Computing (VEC) is a promising technique to improve the quality of service (QoS) and quality of experience (QoE) in autonomous driving by exploiting the resources at the network edge. However, the high mobility of the vehicles leads to stochastic communication link duration, and the tasks generated by various applications in autonomous driving incur fierce competition for resources. These challenges cause excessive task completion delays. In this paper, we propose a vehicle-to-vehicle (V2V) partial computation offloading scheme that leverages the prediction results of the communication link lifetime between vehicles. A History track, Current interactions and Future planning trajectory-aware Gated Recurrent Units (HCF-GRU) network is built to capture the essential factors to improve the prediction accuracy. Then, we design a GRU-based Proximal Policy Optimization (GRU-PPO) algorithm to obtain an optimal one-to-many offloading decision to minimize the task execution cost. The HCF-GRU prediction algorithm is evaluated on a real world vehicle trajectory dataset, and the performance of the GRU-PPO algorithm is analyzed on extensive numerical simulations. Experimental results demonstrate that our prediction network and offloading decision algorithm outperform the baseline methods in terms of prediction accuracy and task execution cost.

Interactive Vehicle Trajectory Prediction for Highways Based on a Graph Attention Mechanism

Precise trajectory prediction is pivotal for autonomous vehicles operating in real-world traffic conditions, and can help them make the right decisions to ensure safety on the road. However, state-of-the-art approaches consider limited information about the historical movements of vehicles. On highways, drivers make their next judgments according to the behavior of the ambient vehicles. Thus, vehicles need to consider temporal and spatial interactions to reduce the risk of future collisions. In the current work, a trajectory prediction method is put forward in accordance with a graph attention mechanism. We add the absolute and relative motion information of vehicles to the input of the model to describe the vehicles’ past motion states more accurately. LSTM models are employed to process the historical motion information of vehicles, as well as the temporal correlations in interactions. The graph attention mechanism is applied to capture the spatial correlations between vehicles. Utilizing a decoder rooted in an LSTM framework, the future trajectory distribution is generated. Evaluation on the NGSIM US-101 and I-80 datasets substantiates the superiority of our approach over existing state-of-the-art algorithms. Moreover, the predictions of our model are analyzed.

Dynamic Modeling, Simulation, and Optimization of Vehicle Electronic Stability Program Algorithm Based on Back Propagation Neural Network and PID Algorithm

Dynamic Modeling, Simulation, and Optimization of Vehicle Electronic Stability Program Algorithm Based on Back Propagation Neural Network and PID Algorithm

Multi‐agent trajectory prediction with adaptive perception‐guided transformers

AbstractThe ability to predict the trajectory of an autonomous vehicle accurately is crucial for safe and efficient navigation. However, predicting diverse and multimodal futures can be challenging. Recent approaches such as attention and graph neural networks have achieved state‐of‐the‐art performance by considering agent interactions and map contexts. This study focused on multi‐agent prediction using an agent‐centric approach with transformers. This enables parallel computation and a comprehensive understanding of the environment. Two main features are introduced: an adaptive receptive field (ARF) that captures the relevant surroundings for each agent, and perception encoding, which serves as spatial context embeddings. The ARF adapts to the agent's velocity and rotation, focusing attention ahead at high speeds or to the sides when it is slower. Perception encoding divides agents or lanes into levels and encodes the information of each level. This approach enables the efficient encoding of complex spatial relationships. The proposed method combines these advances with transformer modelling for multi‐agent trajectory prediction while ensuring real‐time prediction capabilities. The approach is evaluated on the Argoverse benchmark and better performance than the state‐of‐the‐art baseline is achieved. By addressing challenges such as multimodal outputs and robustness, the study enhances the safety and efficiency of autonomous driving systems by more accurately predicting trajectories.

Vehicle group identification and evolutionary analysis using vehicle trajectory data

Vehicles often move forward in groups on the highways, especially when speed and density are high simultaneously. Abnormal maneuvers of a vehicle in a group influence multiple vehicles surrounding it, potentially leading to traffic accidents. We propose an approach to identify vehicle groups and analyse the factors influencing their evolutions using vehicle trajectory data. The proposed approach quantifies the interactions between neighboring vehicles based on the potential energy field, represents the interactive relationships among multiple vehicles using a multi-vehicle interaction network, and adopts the process of sub-network segmentation to identify vehicle groups. A random-parameter logistic regression (RPLG) model is developed to examine the influence of vehicle group features on vehicle group split. The effectiveness of the proposed approach is demonstrated in a case study using a real-world dataset. The case study reveals that: (1) the interaction strengths between vehicles tend to increase with increasing speed, (2) the interaction strength between a vehicle and its preceding vehicle is the largest, while the interaction strengths between a vehicle and its vehicles on its sides are the lowest, and (3) higher longitudinal and lateral speeds, larger fluctuations in longitudinal speeds and accelerations, larger group size, larger distances between vehicles, more lanes occupied by a vehicle group, and higher vehicle interactions significantly increase the probability of vehicle group split. The findings of this study can potentially support the traffic management and development of autonomous driving technology in connected vehicle environments.

Comparative study on fatigue evaluation of suspenders by introducing actual vehicle trajectory data

Suspenders play a crucial role in transmitting loads from the bridge deck to the main cable in a suspension bridge. They are susceptible to fatigue due to repeated dynamic loads, particularly traffic loads. Traffic Load Models (TLMs), typically created using Monte–Carlo simulation and Weigh-In-Motion (WIM) data, are employed to evaluate this fatigue. However, these models often overlook practical vehicle trajectories and spatio-temporal distribution, which compromises the precision of fatigue assessments. In this study, we introduce a novel 2D Intelligent Driver Model (2D-IDM) that incorporates actual vehicle trajectories, with a particular focus on transverse vehicle movement. This enhancement aims to improve the fidelity of existing TLMs. To provide a clear, qualitative, and quantitative understanding of the effects of fatigue evaluation with or without actual trajectory characteristics, we have structured this paper as a comparative study. We compare our proposed model, denoted as TLM S-3, with two observation-based models (O-1 and O-2) and two simulation-based models (S-1 and S-2). We conducted an experimental case study on a long-span suspension bridge, where the actual traffic load trajectory was obtained using a WIM-Vision integrated system. To calculate fatigue damage considering both longitudinal and transverse directions, we established a multi-scale Finite Element Model (FEM) using solid element types to simulate the bridge girder. This model can generate the stress influence surface of the bridge and has been verified in both static and dynamic aspects. Our detailed comparative analysis demonstrates the consistency of the proposed 2D-IDM with the actual measured traffic load trajectories. This indicates that our approach can enhance the fidelity and precision of fatigue evaluations for bridge suspenders.

Comparison at Scale of Traffic Signal Cycle Split Failure Identification from High-Resolution Controller and Connected Vehicle Trajectory Data

Split failures have been a conventional method to estimate overcapacity at signalized intersections. Currently, split failures are estimated from high-resolution (HR) traffic signal controller event data by evaluating occupancy at the stop bar. Recently, a technique that uses high-fidelity connected vehicle (CV) trajectory data to estimate split failures has been developed and has been adopted by some agencies. This paper compares cycle-by-cycle split failure estimations from both techniques for 42 signalized intersections across central Indiana. CV trajectories were assigned to a cycle based on their arrival characteristics. Then, HR and CV data were used to determine whether each cycle split fails. Finally, agreements and discrepancies were quantified and evaluated. The results obtained after analyzing over 35,000 cycles showed that both techniques produce similar overall split failure estimations. The HR and the CV methods identified 4% and 3% of all cycles as split failing, respectively. However, only 23% of all cycles determined as split failing with the HR approach were also identified as split failing with CV data. Similarly, only 30% of all cycles determined as split failing with the CV approach were also identified as split failing with the HR approach. This indicates significant discrepancies on a cycle-by-cycle basis. Using CV data to identify split failing cycles produces more conservative results and is based on the entire experience of traversing vehicles. If data are available, the authors recommend the CV approach when allocating limited agency resources for operational improvement activities.

Predictive Dead Reckoning for Online Peer-to-Peer Games

In online peer-to-peer games, players send periodic updates to each other and each player must locally reconstruct the position of their opponents in between these updates. In scenarios where players are driving cars, high speeds produce more pronounced errors in local replication of online opponents. In this work, we propose a new method of replicating opponents with less data sent and up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$45\%$</tex-math></inline-formula> less error compared to the state-of-the-art. We use a neural network based approach to predict an opponent's position, combined with a path tracking controller from the field of mobile robotics, to produce smooth, believable trajectories for opponents' vehicles. We also propose a neural network based approach to predict a replicated opponent's trajectory following a collision with a static obstacle.

Building a knowledge base for autonomous control of unmanned vehicles

This article presents an approach to building and using a knowledge base for autonomous control of unmanned vehicles (UV). Agriculture is presented as a subject area which features and limitations must be considered. The lack of a sufficient number and level of qualifications of machine operator leads to equipment downtime, a decrease in crop yields and the efficiency of using chemical fertilizers. The use of UV makes it possible to reduce the influence of these factors and the harmful effects on people working in agriculture. The article focuses on taking into account the features and limitations of the subject area when constructing the trajectory of unmanned vehicles and controlling processing facilities. An approach is proposed that consists of separate stages of designing a knowledge base schema, automating the process of filling the knowledge base and organizing the logical inference function. For each stage, developed models and algorithms are presented that help to form and use a knowledge base when solving the problem of autonomous control of unmanned vehicles. The article contains examples and illustrations designed to increase the clarity of the proposed approach.