Search Results (61)

Short-term traffic prediction is vital for intelligent transportation systems, enabling adaptive congestion control, real-time signal management, and dynamic route planning for autonomous vehicles (AVs). This study introduces a custom Transformer-based deep learning framework for joint forecasting of traffic flow and vehicle speed, leveraging handcrafted positional encoding and stacked multi-head attention layers to model multivariate traffic patterns. Evaluated against baselines including Long Short-Term Memory (LSTM), Support Vector Machine (SVM), Random Tree, and Random Forest on the Next-Generation Simulation (NGSIM) dataset, the model achieves 94.2% accuracy (Root Mean Squared Error (RMSE) 0.16) for flow and 92.1% accuracy for speed, outperforming traditional and deep learning approaches. A hybrid evaluation metric, integrating RMSE and threshold-based accuracy tailored to AV operational needs, enhances its practical relevance. With its parallel processing capability, this framework offers a scalable, real-time solution, advancing AV ecosystems and smart mobility infrastructure. Full article

To address the ‘mechanical’ return to original lane and similar non-humanized lane-changing issues that may occur in existing intelligent driving systems after completing overtaking maneuvers, this study proposes a humanized lane-changing decision method that incorporates driving style characteristics. First, based on the NGSIM dataset, we employ cluster analysis to systematically dissect human drivers’ lane-changing behavior patterns, laying the theoretical foundation for constructing a human-like decision framework. Second, a game model is established to precisely represent diverse driving styles by adjusting the weights of safety, efficiency, and comfort objectives. A reference line dynamic switching mechanism is then proposed to optimize lane-change paths by integrating vehicle speed and safety distance. Joint simulation results demonstrate superiority over dynamic programming (DP) methods in multiple aspects: under conservative driving mode, dual safety thresholds for following distance and speed significantly enhance safety and reliability. In general driving mode, driving stability and smoothness improved by 2.64% and 75.28%, respectively; in aggressive driving mode, lane-change speed increased by 7.06%. These improvements demonstrate that the human-like lane-changing strategy can autonomously achieve the optimal dynamic balance between safety, comfort, and efficiency tailored to different driving styles, providing an effective pathway for constructing high-performance autonomous driving decision systems. Full article

Current imitation learning approaches, predominantly based on deep neural networks (DNNs), offer efficient mechanisms for learning driving policies from real-world datasets. However, they suffer from inherent limitations in interpretability and generalizability—issues of critical importance in safety-critical domains such as autonomous driving. In this paper, we introduce Symbolic Imitation Learning (SIL), a novel framework that leverages Inductive Logic Programming (ILP) to derive explainable and generalizable driving policies from synthetic datasets. We evaluate SIL on real-world HighD and NGSim datasets, comparing its performance with state-of-the-art neural imitation learning methods using metrics such as collision rate, lane change efficiency, and average speed. The results indicate that SIL significantly enhances policy transparency while maintaining strong performance across varied driving conditions. These findings highlight the potential of integrating ILP into imitation learning to promote safer and more reliable autonomous systems. Full article

Intelligent Transportation Systems (ITS) play a pivotal role in addressing traffic congestion, inefficiency, and safety concerns. Emergency lane control on highways is a critical ITS component, yet existing strategies often lack flexibility, theoretical rigor, and the ability to handle dynamic spatiotemporal interactions under uncertain data. To address these limitations, this paper introduces Koopman Graph Mamba (KGM), an innovative framework integrating the Koopman operator with a graph-based state space model for dynamic emergency lane control. KGM leverages multimodal traffic data to predict spatiotemporal patterns, facilitating real-time decisions. An interpretable decision module based on fuzzy neural networks ensures context-sensitive decisions. Evaluated on a real-world dataset from the Changshen Expressway (Nanjing-Changzhou section) and public datasets including NGSIM, PeMS04, and PeMS08, KGM demonstrates superior performance with linear computational complexity, underscoring its potential for large-scale, real-time applications. Full article

In urban intelligent transportation systems, the real-time acquisition of network-wide traffic states is constrained by limited sensor density and high deployment costs. To address this challenge, this paper proposes a learnable Detection Point Selection Module (DPSM), which adaptively determines the most informative observation points through an end-to-end attention mechanism to support full-map traffic state estimation. Distinct from conventional fixed deployment strategies, DPSM provides an adaptive detector configuration that, under the same number of loop sensors, achieves significantly higher estimation accuracy by intelligently optimizing their placement. Specifically, the module takes normalized spatial and temporal information as input and generates an attention-based distribution to identify critical traffic flow readings, which are subsequently fed into various backbone prediction models, including fully connected networks, convolutional neural networks, and long short-term memory networks. Experiments on the real-world NGSIM-US101 dataset demonstrate that three variants—DPSM-NN, DPSM-CNN, and DPSM-LSTM—consistently outperform their corresponding baselines, with notable robustness under sparse observation scenarios. These results highlight the advantage of adaptive detector placement in maximizing the utility of limited sensors, effectively mitigating information loss from sparse deployments and offering a cost-efficient, scalable solution for urban traffic monitoring and control. Full article

Deep learning models have achieved significant progress in developing self-driving algorithms. Despite their advantages, these algorithms typically require substantial amounts of data for effective training. Critical driving data, in particular, is essential for enhancing training efficiency and ensuring driving safety. However, existing methods for identifying critical data often rely on human prior knowledge or are disconnected from the training of self-driving algorithms. In this paper, we introduce a novel data distillation technique designed to autonomously identify critical data for training self-driving algorithms. We conducted experiments with both numerical simulations and the NGSIM dataset, which consists of real-world car trajectories on highway US-101, to validate our approach. In the numerical experiments, the distillation method achieved a test root mean squared error of 1.933 using only 200 distilled training data samples, demonstrating a significant improvement in data efficiency compared to the 1.872 test error obtained with 20,000 randomly sampled training samples. The distilled critical data represents only 1% of the original dataset, optimizing data usage and significantly enhancing computational efficiency. For real-world NGSIM data, we demonstrate the performance of the proposed method in scenarios with extremely sparse data availability and show that our proposed data distillation method outperforms other sampling baselines, including Herding and K-centering. These experimental results highlight the capability of the proposed method to autonomously identify critical data without relying on human prior knowledge. Full article

Accurate trajectory prediction of surrounding vehicles is a fundamental challenge in autonomous driving, requiring sophisticated modeling of complex vehicle interactions, traffic dynamics, and contextual dependencies. This paper introduces Adaptive Speed-Aware Prediction (ADSAP), a novel trajectory prediction framework that advances the state of the art through innovative mechanisms for adaptive attention modulation and knowledge transfer. At its core, ADSAP employs an adaptive deformable speed-aware pooling mechanism that dynamically adjusts the model’s attention distribution and receptive field based on instantaneous vehicle states and interaction patterns. This adaptive architecture enables fine-grained modeling of diverse traffic scenarios, from sparse highway conditions to dense urban environments. The framework incorporates a sophisticated speed-aware multi-scale feature aggregation module that systematically combines spatial and temporal information across multiple scales, facilitating comprehensive scene understanding and robust trajectory prediction. To bridge the gap between model complexity and computational efficiency, we propose an adversarial knowledge distillation approach that effectively transfers learned representations and decision-making strategies from a high-capacity teacher model to a lightweight student model. This novel distillation mechanism preserves prediction accuracy while significantly reducing computational overhead, making the framework suitable for real-world deployment. Extensive empirical evaluation on the large-scale NGSIM and highD naturalistic driving datasets demonstrates ADSAP’s superior performance. The ADSAP framework achieves an 18.7% reduction in average displacement error and a 22.4% improvement in final displacement error compared to state-of-the-art methods while maintaining consistent performance across varying traffic densities (0.05–0.85 vehicles/meter) and speed ranges (0–35 m/s). Moreover, ADSAP exhibits robust generalization capabilities across different driving scenarios and weather conditions, with the lightweight student model achieving 95% of the teacher model’s accuracy while offering a 3.2× reduction in inference time. Comprehensive experimental results supported by detailed ablation studies and statistical analyses validate ADSAP’s effectiveness in addressing the trajectory prediction challenge. Our framework provides a novel perspective on integrating adaptive attention mechanisms with efficient knowledge transfer, contributing to the development of more reliable and intelligent autonomous driving systems. Significant improvements in prediction accuracy, computational efficiency, and generalization capability demonstrate ADSAP’s potential ability to advance autonomous driving technology. Full article

In the evolution of autonomous vehicles (AVs), ensuring safety is of the utmost significance. Precise trajectory prediction is indispensable for augmenting vehicle safety and system performance in intricate environments. This study introduces a novel double-layer long short-term memory (LSTM) model to surmount the limitations of conventional prediction methods, which frequently overlook predicted vehicle behavior and interactions. By incorporating driving-style category values and an improved adaptive grid generation method, this model achieves more accurate predictions of vehicle intentions and trajectories. The proposed approach fuses multi-sensor data collected by perception modules to extract vehicle trajectories. By leveraging historical trajectory coordinates and driving style, and by dynamically adjusting grid sizes according to vehicle dimensions and lane markings, this method significantly enhances the representation of vehicle motion features and interactions. The double-layer LSTM module, in conjunction with convolutional layers and a max-pooling layer, effectively extracts temporal and spatial features. Experiments conducted using the Next Generation Simulation (NGSIM) US-101 and I-80 datasets reveal that the proposed model outperforms existing benchmarks, with higher intention accuracy and lower root mean square error (RMSE) over 5 s. The impact of varying sliding window lengths and grid sizes is examined, thereby verifying the model’s stability and effectiveness. Full article

Lane changing is a crucial scenario in traffic environments, and accurately recognizing and predicting lane-changing behavior is essential for ensuring the safety of both autonomous vehicles and drivers. Through considering the multi-vehicle information interaction characteristics in lane-changing behavior for vehicles and the impact of driver experience needs on lane-changing decisions, this paper proposes a lane-changing model for vehicles to achieve safe and comfortable driving. Firstly, a lane-changing intention recognition model incorporating interaction effects was established to obtain the initial lane-changing intention probability of the vehicles. Secondly, by accounting for individual driving styles, a lane-changing behavior decision model was constructed based on a Gaussian mixture hidden Markov model (GMM-HMM) along with a parameter estimation method. The initial lane-changing intention probability serves as the input for the decision model, and the final lane-changing decision is made by comparing the probabilities of lane-changing and non-lane-changing scenarios. Finally, the model was validated using real-world data from the Next Generation Simulation (NGSIM) dataset, with empirical results demonstrating its high accuracy in recognizing and predicting lane-changing behavior. This study provides a robust framework for enhancing lane-changing decision making in complex traffic environments. Full article

The rapid evolution of intelligent vehicle technology has significantly advanced autonomous decision-making and driving safety. However, the challenge of predicting long-term trajectories in complex urban traffic persists, as traditional methodologies usually handle spatiotemporal attention mechanisms in isolation and are typically limited to short-term trajectory predictions. This paper proposes a Habit-based Temporal–Spatial Attention Long Short-Term Memory (HTSA-LSTM) network, a novel framework that integrates a dual spatiotemporal attention mechanism to capture dynamic dependencies across time and space, coupled with a driving style analysis module. The driving style analysis module employs Sparse Inverse Covariance Clustering and Spectral Clustering (SICC-SC) to extract driving primitives and cluster trajectory data, thereby revealing diverse driving behavior patterns without relying on predefined labels. By segmenting real-world driving data into fundamental behavioral units that reflect individual driving preferences, this approach enhances the model’s adaptability. These behavioral units, in conjunction with the spatiotemporal attention outputs, serve as inputs to the model, ultimately improving prediction accuracy and robustness in multi-vehicle scenarios. The model was evaluated by using the NGSIM dataset and real driving data from Wuhan, China. In comparison to benchmark models, HTSA-LSTM achieved a 20.72% reduction in the root mean square error (RMSE) and a 24.98% reduction in the negative log likelihood (NLL) for 5 s predictions of long-term trajectories. Furthermore, HTSA-LSTM achieved R² values exceeding 97.9% for 5 s predictions on highways and expressways and over 92.7% for 3 s predictions on urban roads, highlighting its excellent performance in long-term trajectory prediction and adaptability across diverse driving conditions. Full article

In order to quickly and accurately identify the lane changing intention of vehicles, and to deeply consider the time series characteristics of vehicle driving processes and the interactive effects between vehicles, a lane changing intention recognition model, namely, Model_TA, was constructed by combining the time series feature extraction ability of the encoder in the Transformer model, the bidirectional gating mechanism of the bidirectional gated recurrent unit, and the additive attention mechanism. The performance of the Model_TA model was trained and validated on the I-80 dataset in NGSIM. The experimental results showed that the accuracy of model intent recognition was 97.01%, which was 20.3%, 4.73%, and 1.73% higher than that of SVM, LSTM, and Transformer models, respectively; the prediction accuracy at 2.0 s, 2.5 s, and 3.0 s is 90.15%, 84.58%, and 83.13%, respectively, which is better than similar models. It is proved that the model can better predict the lane changing intention of vehicles. Full article

Accurately predicting the future trajectory of road users around autonomous vehicles is crucial for path planning and collision avoidance. In recent years, data-driven vehicle trajectory prediction models have become a significant research focus, and various spatial–temporal neural network models, based on spatial–temporal data, have been proposed. However, some existing spatial–temporal models segregate time and space, neglecting the inherent coupling of time and space. To address this issue, an end-to-end spatial–temporal feature fusion model, based on the Vision Transformer (Vit), is proposed in this paper, which can couple stereoscopic features of diverse spatial regions and time periods. Specifically, we propose an end-to-end spatiotemporal feature coupling model based on visual Transformer, Vit-Traj, which extracts spatiotemporal features through 2D convolution and uses Vit and SENet to complete feature fusion. Experimental results on the NGSIM and HighD datasets indicate that, compared to State-of-the-Art models, the proposed model exhibits better performance. The root mean squared error (RMSE) is 2.72 m on the NGSIM dataset and 0.86 m on the HighD dataset when the prediction horizon is 5 s. Furthermore, ablation experiments are conducted to evaluate the performance of each module, affirming the efficacy of ViT in modeling spatial–temporal data. Full article

Due to advances in sensor techniques and deep learning, autonomous vehicular technologies have become more reliable and practical. Trajectory prediction is a critical task to anticipate the future positions of surrounding vehicles. However, existing algorithms, such as LSTM-based and attention-based models, face challenges of high computational complexity, large parameter sizes, and limited ability to efficiently capture both temporal dependencies and spatial interactions in dynamic traffic scenarios. In this paper, we propose a parameter-efficient trajectory prediction model that integrates Liquid Time-Constant (LTC) networks with attention mechanisms, termed the Attn-LTC model. The key contributions of our work are threefold. First, we introduce a temporal attention-enhanced LTC encoder that effectively captures both long-term temporal dependencies and dynamic behaviors from historical trajectory data. Second, we incorporate a spatial attention-enhanced LTC decoder, which emphasizes the influence of neighboring vehicles and spatial interactions, thereby improving prediction accuracy. Third, we demonstrate the computational efficiency of the Attn-LTC model, which achieves high predictive accuracy with significantly fewer parameters compared to LSTM-based and Transformer-based counterparts. Extensive experiments conducted on the NGSIM dataset demonstrate the advantages of our proposed Attn-LTC model. Notably, it reduces computational complexity and model size while maintaining superior accuracy, making it well suited for deployment in resource-constrained systems. The results highlight the effectiveness of the Attn-LTC model in balancing precision and efficiency, paving the way for its application in real-time autonomous driving systems. Full article

At present, achieving efficient, sustainable, and safe transportation has led to increasing attention on driving behavior recognition and advancements in autonomous driving. Identifying diverse driving styles and corresponding types is crucial for providing targeted training and assistance to drivers, enhancing safety awareness, optimizing driving costs, and improving autonomous driving systems responses. However, current studies mainly focus on specific driving scenarios, such as free driving, car-following, and lane-changing, lacking a comprehensive and systematic framework to identify the diverse driving styles. This study proposes a novel, data-driven approach to driving-style recognition utilizing naturalistic driving data NGSIM. Specifically, the NGSIM dataset is employed to categorize car-following and lane-changing groups according to driving-state extraction conditions. Then, characteristic parameters that fully represent driving styles are optimized through correlation analysis and principal component analysis for dimensionality reduction. The K-means clustering algorithm is applied to categorize the car-following and lane-changing groups into three driving styles: conservative, moderate, and radical. Based on the clustering results, a comprehensive evaluation of the driving styles is conducted. Finally, a comparative evaluation of SVM, Random Forest, and KNN recognition indicates the superiority of the SVM algorithm and highlights the effectiveness of dimensionality reduction in optimizing characteristic parameters. The proposed method achieves over 97% accuracy in identifying car-following and lane-changing behaviors, confirming that the approach based on naturalistic driving data can effectively and intelligently recognize driving styles. Full article

In automatic driving, accurate prediction of vehicle trajectory is the key to achieve automatic driving, and multi-vehicle joint trajectory prediction has become an important part of modern human-computer interaction systems such as automatic driving. In order to better predict vehicle trajectories, we propose a new residual diffusion model to infer the joint distribution of future multi-vehicle trajectories. This approach has several major advantages. First, the model is able to learn multiple probability distributions from trajectory data to obtain potential outcomes for vehicles to multiple future trajectories. Secondly, in order to integrate the motion characteristics of multiple vehicles in the same scene, we use the method of combining the reference denoising and multiple residual denoising to improve the model performance and prediction speed. Finally, on this basis, a general trajectory constraint function is introduced, so that the generated trajectories of multiple vehicles will not collide with each other. We perform a rich experimental comparison of various existing methods on the NGSIM dataset and demonstrate that the proposed algorithm achieves a 26% improvement on $mAP$. Full article