For Autonomous Vehicles (AVs), recognizing traffic lights and signs is critical for safety because perception errors directly affect navigation decisions. Real-world disturbances such as glare, rain, dirt, and graffiti, as well as digital adversarial attacks, can lead to dangerous misclassifications. Current research lacks (i) temporal continuity (stable detection across consecutive frames to prevent flickering misclassifications), (ii) multi-field-of-view (FoV) sensing, and (iii) integrated defenses against both digital and natural degradation. This paper presents two principal contributions: (1) a three-layer defense framework integrating feature squeezing, inference-time temperature scaling (softmax τ = 3 without distillation training), and entropy-based anomaly detection with sequence-level temporal voting; (2) a 500 sequence dual-FoV benchmark (30k base frames, 150k with perturbations) from aiMotive, Waymo, Udacity, and Texas sources across four operational design domains. The unified defense stack achieves 79.8% mAP on a 100-sequence test set (6k base frames, 30k with perturbations), reducing attack success rate from 37.4% to 18.2% (51% reduction) and high-risk misclassifications by 32%. Cross-FoV validation and temporal voting enhance stability under lighting changes (+3.5% mAP) and occlusions (+2.7% mAP). Defense improvements (+9.5-9.6% mAP) remain consistent across native 3D (aiMotive, Waymo) and projected 2D (Udacity, Texas) annotations. Preliminary recapture experiments (n = 15 scenarios) show 2.5% synthetic-physical ASR gap (p = 0.18), though larger validation is needed. Code, models, and dataset reconstruction tools are publicly available.
LiDAR-based multi-task perception for autonomous driving requires efficient integration of spatial context and cross-task information, yet existing methods often suffer from restricted receptive fields and suboptimal task interaction. This paper presents a novel multi-task framework that goes beyond sparsity constraints, leveraging receptive field expansion and cross-task fusion to enhance 3D object detection and semantic segmentation. We introduce the Spatial Density-Invariant Multi-scale Integrator (SDIMI), which adaptively fuses multi-resolution contextual features using density-agnostic strategies to expand the receptive field while preserving feature sparsity. Additionally, the Synergistic Instance-Driven Multitask Fusion (SIDMF) module dynamically aligns instance-level features between segmentation and detection, enabling efficient high-level feature propagation via bounding box mapping masks. Experiments on NuScenes and Waymo Open Dataset demonstrate state-of-the-art performance: our method achieves 72.0% NuScenes Detection Score (NDS) and 84.4% mIOU on NuScenes, and 79.8% mAPH-L2 and 72.3% mIOU on Waymo.
Camera-based 3D object detection in BEV (Bird's Eye View) space has drawn great attention over the past few years. Dense detectors typically follow a two-stage pipeline by first constructing a dense BEV feature and then performing object detection in BEV space, which suffers from complex view transformations and high computation costs. On the other side, sparse detectors follow a query-based paradigm without explicit dense BEV feature construction but generally underperform compared to dense ones. In this paper, we find that the key to mitigating this performance gap is the adaptability of the detector in both BEV and image space. To this end, we propose a fully sparse 3D object detector that outperforms the dense counterparts and enjoys a higher running speed. Our sparse detector contains three key designs, which are (1) scale-adaptive self attention to aggregate features with adaptive receptive field in BEV space, (2) scale-adaptive cross attention to capture the unique temporal dynamics associated with different objects, (3) adaptive sampling and mixing to perform interactions between queries and image features under the guidance of queries. These key components enhance the adaptability of the detector in both BEV and image space. Furthermore, we explore two distinct temporal modeling approaches: sampling-point-based multi-frame stacking (dubbed SparseBEV) and query-based recurrent temporal fusion (dubbed SparseBEV++) to leverage temporal features effectively. Experiments are conducted on the nuScenes and Waymo datasets. On the val split of nuScenes, both SparseBEV and SparseBEV++ surpass all previous methods. Our SparseBEV achieves a performance of 55.8 NDS and a speed of 23.5 FPS, and SparseBEV++ further achieves a remarkable 57.1 NDS while maintaining a real-time inference speed of 24.6 FPS. On the Waymo dataset, our best-performing model, SparseBEV++, outperforms previous methods with a lead of 58.9 mAP and 55.2 mAPH.
High-fidelity 4D reconstruction of dynamic scenes is pivotal for immersive simulation yet remains challenging due to the photometric inconsistencies inherent in multi-view sensor arrays. Standard 3D Gaussian Splatting (3DGS) strictly adheres to the brightness constancy assumption, failing to distinguish between intrinsic scene radiance and transient brightness shifts caused by independent auto-exposure (AE), auto-white-balance (AWB), and non-linear ISP processing. This misalignment often forces the optimization process to compensate for spectral discrepancies through incorrect geometric deformation, resulting in severe temporal flickering and spatial floating artifacts. To address these limitations, we present Lumina-4DGS, a robust framework that harmonizes spatiotemporal geometry modeling with a hierarchical exposure compensation strategy. Our approach explicitly decouples photometric variations into two levels: a Global Exposure Affine Module that neutralizes sensor-specific AE/AWB fluctuations and a Multi-Scale Bilateral Grid that residually corrects spatially varying non-linearities, such as vignetting, using luminance-based guidance. Crucially, to prevent these powerful appearance modules from masking geometric flaws, we introduce a novel SSIM-Gated Optimization mechanism. This strategy dynamically gates the gradient flow to the exposure modules based on structural similarity. By ensuring that photometric enhancement is only activated when the underlying geometry is structurally reliable, we effectively prioritize geometric accuracy over photometric overfitting. Extensive experiments validate the quantitative superiority of Lumina-4DGS. On the Waymo Open Dataset, our method achieves a state-of-the-art Full Image PSNR of 31.12 dB while minimizing geometric errors to a Depth RMSE of 1.89 m and Chamfer Distance of 0.215 m. Furthermore, on our highly challenging self-collected surround-view dataset featuring severe unconstrained illumination shifts, Lumina-4DGS yields a significant 2.13 dB PSNR improvement over recent driving-scene baselines. These results confirm that our framework achieves photorealistic, exposure-invariant novel view synthesis while maintaining superior geometric consistency across heterogeneous camera inputs.
Adversarial attack strategies for 3D object detection have highlighted the critical importance of addressing security concerns in this domain. However, white-box methods require full access to the victim model in large-scale point cloud applications. To this end, we propose a novel Policy-Driven Black-box Attack (BAT) that is designed to optimize attack locations without necessitating detailed knowledge of the victim models. First, we introduce a density-aware pattern generator that creates scene-adaptive attack clusters. Second, we leverage the deep deterministic policy gradient in deep reinforcement learning to train an attack agent capable of targeting the victim model. Ultimately, the attack agent is iteratively directed towards optimal attack locations through the joint application of critic loss and actor loss. To the best of our knowledge, this represents the first reinforcement learning-based black-box attack applied to practical 3D object detection. Experimental results on the KITTI, nuScenes, and Waymo datasets demonstrate that BAT effectively diminishes the accuracy of notable models. Importantly, BAT significantly enhances the attack success rate (surpassing state-of-the-art both white-box and black-box methods) and increases transferability (by 20 times) through simple deep deterministic policy gradient, thus establishing a new baseline for adversarial attacks in 3D object detection.
Three-dimensional reconstruction in scenes with extreme depth variations remains challenging due to inconsistent supervisory signals between near-field and far-field regions. Existing methods fail to simultaneously address inaccurate depth estimation in distant areas and structural degradation in close-range regions. This paper proposes a novel computational framework that integrates depth-of-field supervision and multi-view consistency supervision to advance 3D Gaussian Splatting. Our approach comprises two core components: (1) Depth-of-field Supervision employs a scale-recovered monocular depth estimator (e.g., Metric3D) to generate depth priors, leverages defocus convolution to synthesize physically accurate defocused images, and enforces geometric consistency through a novel depth-of-field loss, thereby enhancing depth fidelity in both far-field and near-field regions; (2) Multi-View Consistency Supervision employing LoFTR-based semi-dense feature matching to minimize cross-view geometric errors and enforce depth consistency via least squares optimization of reliable matched points. By unifying defocus physics with multi-view geometric constraints, our method achieves superior depth fidelity, demonstrating a 0.8 dB PSNR improvement over the state-of-the-art method on the Waymo Open Dataset. This framework bridges physical imaging principles and learning-based depth regularization, offering a scalable solution for complex depth stratification in urban environments.
Recent years have witnessed the remarkable progress of 3D multi-modality object detection methods based on the Bird's-Eye-View (BEV) perspective. However, most of them overlook the complementary interaction and guidance between LiDAR and camera. In this work, we propose a novel multi-modality 3D objection detection method, with multi-guided global interaction and LiDAR-guided adaptive fusion, named MGAF. Specifically, we introduce sparse depth guidance (SDG) and LiDAR occupancy guidance (LOG) to generate 3D features with sufficient depth and spatial information. The designed semantic segmentation network captures category and orientation prior information for raw point clouds. In the following, an Adaptive Fusion Dual Transformer (AFDT) is developed to adaptively enhance the interaction of different modal BEV features from both global and bidirectional perspectives. Meanwhile, additional downsampling with sparse height compression and multi-scale dual-path transformer (MSDPT) are designed in order to enlarge the receptive fields of different modal features. Finally, a temporal fusion module is introduced to aggregate features from previous frames. Notably, the proposed AFDT is general, which also shows superior performance on other models. Our framework has undergone extensive experimentation on the large-scale nuScenes dataset, Waymo Open Dataset, and long-range Argoverse2 dataset, consistently demonstrating state-of-the-art performance.
Large-scale terrestrial laser scanning (TLS) point clouds are increasingly used for applications such as digital twins and cultural heritage documentation; however, removing unwanted human points captured during acquisition remains a largely manual and time-consuming process. This study proposes a geometry-aware framework for automatically removing human noise from TLS point clouds by projecting 2D instance segmentation masks (obtained using You Only Look Once (YOLO) v8 with an instance segmentation head) into 3D space and validating candidates through multi-stage geometric filtering. To suppress false positives induced by reprojection misalignment and planar background structures (e.g., walls and ground), we introduce projection-followed geometric validation (or "geometric gating") using Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and principal component analysis (PCA)-based planarity analysis, followed by cluster-level plausibility checks. Experiments were conducted on two real-world outdoor TLS datasets-(i) Osaka Metropolitan University Sugimoto Campus (OMU) (82 scenes) and (ii) Jinaimachi historic district in Tondabayashi (JM) (68 scenes). The results demonstrate that the proposed method achieves high noise removal accuracy, obtaining precision/recall/intersection over union (IoU) of 0.9502/0.9014/0.8607 on OMU and 0.8912/0.9028/0.8132 on JM. Additional experiments on mobile mapping system (MMS) data from the Waymo Open Dataset demonstrate stable performance without parameter recalibration. Furthermore, quantitative and qualitative comparisons with representative time-series geometric dynamic object removal methods, including DUFOMap and BeautyMap, show that the proposed approach maintains competitive recall under a human-only ground-truth definition while reducing over-removal of static structures in TLS scenes, particularly when humans are observed in only one or a few scans due to limited revisit frequency. The end-to-end processing time with YOLOv8 was 935.62 s for 82 scenes (11.4 s/scene) on OMU and 571.58 s for 68 scenes (8.4 s/scene) on JM, supporting practical efficiency on high-resolution TLS imagery. Ablation studies further clarify the role of each stage and indicate stable performance under the observed reprojection errors. The annotated human point cloud dataset used in this study has been publicly released to facilitate reproducibility and further research on human noise removal in large-scale TLS scenes.
3D scene flow represents the dense per-point motion field in dynamic scenes, playing a crucial role in various downstream tasks, including motion segmentation, dynamic scene reconstruction, 4D content generation, etc. However, previous regression-based works commonly suffer from unreliable correlations caused by locally constrained search ranges and struggle with the absence of timely feedback regarding the flow estimation uncertainty during training. To address these challenges, we propose a novel uncertainty-aware network for scene flow estimation, termed DifFlow3D, based on the conditional probabilistic diffusion model. Hierarchical diffusion-based flow estimation blocks are designed to enhance the correlation robustness and resilience to challenging cases, e.g., dynamics, noisy inputs, repetitive patterns, etc. To mitigate the generation diversity, three key flow-related features are leveraged as conditions in our diffusion model. Furthermore, we develop an uncertainty estimation module within diffusion to assess the reliability of estimated scene flow dynamically. A Hidden State Denoising strategy (HSD) is also introduced to further boost the stability of the reverse denoising process. Extensive experiments conducted on four scene flow datasets, including both synthetic and real-world datasets (FlyingThings3D, KITTI 2015, Argoverse, and Waymo Open), demonstrate the superiority of our proposed DifFlow3D. Compared to prior state-of-the-art methods, DifFlow3D has 26.0%, 36.4%, 35.3%, and 17.7% EPE3D reduction respectively across four datasets. Only trained on the synthetic FlyingThings3D dataset, our method achieves an unprecedented millimeter-level accuracy (0.0070 m EPE3D) on the real-scene KITTI dataset, highlighting its exceptional generalization capability. Additionally, our diffusion-based refinement paradigm can be seamlessly integrated as a plug-and-play module into existing scene flow networks, significantly enhancing their estimation accuracy. We also introduce our pre-trained scene flow estimator as explicit motion priors into the novel dynamic LiDAR view synthesis task, which validates its great potential for improving the 4D LiDAR reconstruction performance.
In recent years, with the development of autonomous driving, 3D reconstruction for unbounded large-scale scenes has attracted researchers' attention. Existing methods have achieved outstanding reconstruction accuracy in autonomous driving scenes, but most of them lack the ability to edit scenes. Although some methods have the capability to edit scenarios, they are highly dependent on manually annotated 3D bounding boxes, leading to their poor scalability. To address the issues, we introduce a new Gaussian representation, called DrivingEditor, which decouples the scene into two parts and handles them by separate branches to individually model the dynamic foreground objects and the static background during the training process. By proposing a framework for decoupled modeling of scenarios, we can achieve accurate editing of any dynamic target, such as dynamic objects removal, adding and etc, meanwhile improving the reconstruction quality of autonomous driving scenes especially the dynamic foreground objects, without resorting to 3D bounding boxes. Extensive experiments on Waymo Open Dataset and KITTI benchmarks demonstrate the performance in 3D reconstruction for both dynamic and static scenes. Besides, we conduct extra experiments on unstructured large-scale scenarios, which can more convincingly demonstrate the performance and robustness of our proposed model when rendering the unstructured scenes. Our code is available at https://github.com/WangXu-xxx/DrivingEditor.
Autonomous vehicles hold great promise for reducing traffic fatalities and improving transportation efficiency, yet their widespread adoption hinges on embedding credible and transparent ethical reasoning into routine and emergency maneuvers, particularly to protect vulnerable road users (VRUs) such as pedestrians and cyclists. Here, we present a hierarchical Safe Reinforcement Learning (Safe RL) framework that augments standard driving objectives with ethics-aware cost signals. At the decision level, a Safe RL agent is trained using a composite ethical risk cost, combining collision probability and harm severity, to generate high-level motion targets. To improve sample efficiency under rare high-risk events, we introduce a risk-sensitive prioritized experience replay scheme. We further propose Temporal Cost Aggregation (TCA), which propagates risk across decision steps and aligns learning with tail-risk measures such as Conditional Value-at-Risk (CVaR), mitigating single-step independence assumptions. At the execution level, polynomial trajectory generation coupled with Proportional-Integral-Derivative (PID) and Stanley controllers ensures smooth and feasible execution. We evaluate EthicAR in closed-loop simulations based on the Waymo Open Dataset across 75 real-world scenarios and five random seeds. The proposed method decreases collision rates by 20$\sim$45% compared to baseline methods, while maintaining task success rates and comfort metrics within 5$\sim$10% of baselines. This work provides a reproducible benchmark for Safe RL with explicitly ethics-aware objectives in human-mixed traffic scenarios. Our results highlight the potential of combining formal control theory and data-driven learning to advance ethically accountable autonomy that explicitly protects those most at risk in urban traffic environments.
In the past few decades, autonomous driving algorithms have made significant progress in perception, planning, and control. However, evaluating individual components does not fully reflect the performance of entire systems, highlighting the need for more holistic assessment methods. This motivates the development of HUGSIM, a closed-loop, photo-realistic, and real-time simulator for evaluating autonomous driving algorithms. We achieve this by lifting captured 2D RGB images into the 3D space via 3D Gaussian Splatting, improving the rendering quality for closed-loop scenarios, and building the closed-loop environment. In terms of rendering, we tackle challenges of novel view synthesis in closed-loop scenarios, including viewpoint extrapolation and 360-degree vehicle rendering. Beyond novel view synthesis, HUGSIM further enables the full closed simulation loop, dynamically updating the ego and actor states and observations based on control commands. Moreover, HUGSIM offers a comprehensive benchmark across more than 70 sequences from KITTI-360, Waymo, nuScenes, and PandaSet, along with over 400 varying scenarios, providing a fair and realistic evaluation platform for existing autonomous driving algorithms. HUGSIM not only serves as an intuitive evaluation benchmark but also unlocks the potential for fine-tuning autonomous driving algorithms in a photorealistic closed-loop setting.
This paper aims to tackle the problem of modeling dynamic urban streets for autonomous driving scenes. Recent methods extend NeRF by incorporating tracked vehicle poses to animate vehicles, enabling photo-realistic view synthesis of dynamic urban street scenes. However, significant limitations are their slow training and rendering speed. We introduce Street Gaussians, a new explicit scene representation that tackles these limitations. Specifically, the dynamic urban scene is represented as a set of point clouds equipped with semantic logits and Gaussian primitives, each associated with either a foreground object or the background. To model the dynamics of foreground objects, each object point cloud is optimized with optimizable tracked poses, along with a 4D spherical harmonics model for the dynamic appearance. The explicit representation allows easy composition of objects and background, which in turn allows for scene editing operations and rendering at 135 FPS (1066 * 1600 resolution) within half an hour of training. The proposed method is evaluated on multiple challenging benchmarks, including KITTI and Waymo Open datasets. Experiments show that the proposed method consistently outperforms state-of-the-art methods across all datasets.
Merging is one of the key maneuvers where autonomous vehicles (AVs) can perform significantly better in decision-making and execution than human-driven vehicles (HDVs). However, past studies have investigated AV merging in simulation environments but have not comprehensively analyzed the interactions between AVs and HDVs during merging events using real-world merging event datasets. This study examines AV merging behavior in mixed traffic environments by extracting and analyzing merging events from the Argoverse-2 and Waymo AV testing datasets. A Weibull random parameter hazard-based duration model was developed to examine the effect of different driving volatilities and traffic measures (i.e., relative speed, velocity standard deviation, and merging location) on merging Gap Time (GT), representing the aggressiveness and merging event crash risk. Descriptive analysis of the AV and HDV merging events showed that the merging GT distributions for AVs and HDVs were similar. Higher GT reduced the following vehicle's speed variation at the target lane for both AV and HDV merging events. Merging crash risk was estimated using the extreme value theory (EVT) approach. The crash risk analysis revealed similar crash risks irrespective of the presence of AVs in merging events. However, AVs were associated with a reduced severity of crashes/conflicts, suggesting that the programmed behavior of AVs, such as smoother acceleration and adherence to traffic rules, contributed to safer merging in the mixed traffic environment. This study's findings identified a need for improvement in AV technology to ensure safer and more efficient operations in mixed traffic. Merging event safety improvements can be achieved by optimizing AV algorithms for dynamic and human-centric traffic conditions during complex driving scenarios.
With the rapid development of Automated Vehicle (AV) technologies, a mixed traffic environment comprising AVs and Human-Driven Vehicles (HVs) is expected to persist over the long term. While current research primarily focuses on the characteristics of car-following behavior between AVs and HVs, studies addressing conflict risk in these behaviors remain relatively limited. Using the Waymo Open Dataset for autonomous driving, this study empirically evaluates various advanced random parameter frameworks to investigate the influencing factors of conflict risk in three different car-following scenarios: AV-HV, HV-AV, and HV-HV, within the mixed traffic environment. Specifically, the Rear-end Collision Risk Index (RCRI) is defined as the binary outcome variable based on longitudinal car-following behavior data, with explanatory variables including kinematic variables of the preceding and following vehicles, inter-vehicle interactions, and environmental variables. The study primarily examines the suitability of model selection, explores unobserved multilayer heterogeneity, and compares the key factors influencing conflict risk across various car-following scenarios. The results indicate that in the three car-following scenarios-AV-HV, HV-AV, and HV-HV-the optimal models are the random parameters multinomial logit model (RPL), the random parameters multinomial logit model with heterogeneity in means and variances (RPLHMV), and the correlated random parameters multinomial logit model with heterogeneity in means (CRPLHM), respectively. The estimates derived from these optimal models reveal the random parameters, their heterogeneity in means and variances, and the potential correlations among the factors influencing conflict risks. This effectively captures the complex interactions between multiple factors, thereby reducing estimation biases. Furthermore, the significant factors and their respective magnitudes of impact on conflict risks vary across the three car-following scenarios. These variations across different car-following behaviors can enhance the accuracy of behavioral modeling and micro-simulation in mixed traffic flow, and support the formulation of differentiated traffic safety improvement measures.
Recent advancements in radiance fields, particularly with the emergence of Gaussian splatting, have highlighted their significant potential for 3D scene reconstruction and novel view synthesis. However, existing methods encounter substantial challenges when addressing dynamic environments, especially in complex urban settings with both rigid and non-rigid participants. To tackle these challenges, we propose a geometry-aware framework that integrates Gaussian primitives with a template mesh to effectively represent dynamic objects. This integration facilitates the efficient and accurate reconstruction of urban scenes, ensuring that the geometric integrity of dynamic elements is maintained. We first decompose the scene into a dynamic scene graph and fit the template vertices to observations to construct topologically consistent 3D models. Then, we build Gaussian radiance fields for dynamic nodes based on the template meshes, optimizing the vertex offset of dynamic participants to align with their geometric surfaces. We further project the appearance attributes into the 2D texture space based on topological relationships preserved in the Gaussians, enabling finer reconstruction of small-scale details and smoother appearance generalization on unseen surfaces. To validate the effectiveness of our proposed method, we conduct extensive evaluations on the Waymo Open Dataset (Ettinger et al., 2021) and the KITTI Dataset (Geiger et al., 2013). Our results demonstrate superior performance compared to mainstream dynamic reconstruction methods. We believe our work establishes a foundation for more realistic and geometrically complete urban scene reconstruction.
This paper outlines a systematic approach to tackle the creation of a safety case for Automated Driving Systems (ADS) that operate without a driver. A safety case is a formal way to explain how an ADS developer determines that its system is safe enough to be deployed on public roads without a human driver, and it includes evidence to support that determination. It involves an explanation of the system, the methodologies used to develop it, the metrics used to validate it and the actual results of validation tests. Yet, in order to develop a worthwhile safety case, it is first important to understand what makes it credible and well crafted, and align on evaluation criteria. This paper helps enable such alignment by providing foundational thinking into not only how a system is determined to be ready for deployment but also into justifying that the set of acceptance criteria employed in such determination is sufficient and that their evaluation (and associated methods) is credible. The presentation is anchored around the acknowledgement that absolute zero risk is unattainable, framing the definition of safety around the notion of "absence of unreasonable risk" in accordance with state of the art safety standards. The publication is structured around three complementary perspectives on safety: a layered approach to safety; a dynamic approach to safety; and a credible approach to safety. Each perspective focuses on the principles and methodological approach, rather than specific results that are often proprietary and this paper does not feature a full safety case nor the evidence to support it. While centered around the example of a SAE Level 4 ADS, the proposed approach is technology- and methodology-agnostic, making it adaptable for use in whole or in part by any entity in the field.
Field data analysis has shown that SUVs and pickup trucks cause more torso injuries than sedans, and the rapid increasing proportion of SUVs among the U.S. vehicle fleet will likely increase the importance of pedestrian torso protection. The objective of this study is to use finite element (FE) vehicle and human body models to investigate effects of vehicle front-end geometry and stiffness characteristics on pedestrian injuries, specifically focusing on SUVs and pickup trucks and pedestrian torso injuries. Front-end geometries of 74 U.S. vehicles, including 41 with hood leading edge (HLE) > 1000 mm and 33 with 750 mm < HLE < 1000 mm, were collected and analyzed using principal component analysis (PCA). The resulting parametric vehicle front-end geometry model was then linked to an FE generic vehicle (GV) model, so that the GV model can be morphed into a wide range of vehicle front-end geometries representing the fleet. Impact simulations using GHBMC F05, M50, and M95 pedestrian models and three detailed vehicle FE models were conducted with the pedestrian perpendicular to the vehicle front-end located at the center of the vehicle. These simulation results were used to calibrate the stiffness values and contact definitions of the hood and hood leading edge components of the morphed GV models. After GV model calibration, several parametric studies were conducted, resulting in a total of 306 vehicle-to-pedestrian crash simulations using 34 morphed GV models with varied front-end geometric and stiffness characteristics and three pedestrian models under three impact velocities (30, 40, and 50 kph). Pedestrian torso injuries were measured by lateral torso deflections at 17 locations across the chest and abdomen regions. Multiple regression was used to test the significance of the variables. PCA results showed that the top three principal components (PCs) captured over 90% of the variation in vehicle front-end geometries, primarily reflecting HLE height/length, HLE roundness, and overall front-end shape. Simulation results suggested that HLE height and impact velocity were the two dominant factors influencing pedestrian torso injury predictions. Torso injury metrics were the highest when the HLE height was equal to or slightly lower than (<150 mm) the pedestrian's mid-sternum height. In addition, increased HLE roundness and a more compliant HLE were associated with reduced pedestrian torso injuries. This study generated a comprehensive set of vehicle-to-pedestrian impact simulation data, enabling a systematic evaluation of how vehicle front-end geometric and stiffness characteristics influence pedestrian torso injuries.
Collision avoidance - involving a rapid threat detection and quick execution of the appropriate evasive maneuver - is a critical aspect of driving. However, existing models of human collision avoidance behavior are fragmented, focusing on specific scenarios or only describing certain aspects of the avoidance behavior, such as response times. This paper addresses these gaps by proposing a computational cognitive model of human collision avoidance behavior based on active inference. Active inference provides a unified approach to modeling human behavior: the minimization of free energy. Building on prior active inference work, our model incorporates established cognitive mechanisms such as evidence accumulation to simulate human responses in three distinct collision avoidance scenarios: front-to-rear lead vehicle braking, lateral incursion by an oncoming vehicle, and another vehicle failing to yield at an intersection. We demonstrate that our model explains a wide range of empirical findings on human collision avoidance behavior. Specifically, the model closely reproduces both aggregate results from meta-analyses previously reported in the literature and detailed, scenario-specific effects observed in two recent driving simulator studies, including response timing, maneuver selection, and execution. Our results highlight the potential of active inference as a generalizable framework for understanding and modeling human behavior in complex real-life driving tasks.
Autonomous driving simulation system plays a crucial role in enhancing self-driving data and simulating complex and rare traffic scenarios, ensuring navigation safety. However, traditional simulation systems, which often heavily rely on manual modeling and 2D image editing, struggled with scaling to extensive scenes and generating realistic simulation data. In this study, we present S-NeRF++, an innovative autonomous driving simulation system based on neural reconstruction. Trained on widely-used self-driving datasets, such as nuScenes and Waymo, S-NeRF++ can generate a large number of realistic street scenes and foreground objects with high rendering quality as well as offering considerable flexibility in manipulation and simulation. Specifically, S-NeRF++ is an enhanced neural radiance field for synthesizing large-scale scenes and moving vehicles, with improved scene parameterization and camera pose learning. The system effectively utilizes noisy and sparse LiDAR data to refine training and address depth outliers, ensuring high-quality reconstruction and novel-view rendering. It also provides a diverse foreground asset bank by reconstructing and generating different foreground vehicles to support comprehensive scenario creation. Moreover, we have developed an advanced foreground-background fusion pipeline that skillfully integrates illumination and shadow effects, further enhancing the realism of our simulations. With the high-quality simulated data provided by our S-NeRF++, we found the perception methods enjoy performance boosts on several autonomous driving downstream tasks, further demonstrating our proposed simulator's effectiveness.