While much research has recently focused on generating physics-based adversarial samples, a critical yet often overlooked category originates from physical failures within on-board cameras—components essential to the perception systems of autonomous vehicles. Camera failures, whether due to external stresses causing hardware breakdown or internal component faults, can directly jeopardize the safety and reliability of autonomous driving systems. Firstly, we motivate the study using two separate real-world experiments to showcase that indeed glass failures would cause the detection based neural network models to fail. Secondly, we develop a simulation-based study using the physical process of the glass breakage to create perturbed scenarios, representing a realistic class of physics-based adversarial samples. Using a finite element model (FEM)-based approach, we generate surface cracks on the camera image by applying a stress field defined by particles within a triangular mesh. Lastly, we use physically-based rendering (PBR) techniques to provide realistic visualizations of these physically plausible fractures. To assess the safety implications, we apply the simulated broken glass effects as image filters to two autonomous driving datasets- KITTI and BDD100K- as well as the large-scale image detection dataset MS-COCO. We then evaluate detection failure rates for critical object classes using CNN-based object detection models (YOLOv8 and Faster R-CNN) and a transformer-based architecture with Pyramid Vision Transformers. To further investigate the distributional impact of these visual distortions, we compute the Kullback-Leibler (K-L) divergence between three distinct data distributions, applying various broken glass filters to a custom dataset (captured through a cracked windshield), as well as the KITTI and Kaggle cats and dogs datasets. The K-L divergence analysis suggests that these broken glass filters do not introduce significant distributional shifts. Our goal is to provide a robust, physics-based methodology for generating adversarial samples that reflect real-world camera failures, with the overarching aim of improving the resilience and safety of autonomous driving systems against such physical threats.

Subgraph isomorphism, also known as subgraph matching, is typically regarded as an NP-complete problem. This complexity is further compounded in practical applications where edge weights are real-valued and may be affected by measurement noise and potential missing data. Such graph matching routinely arises in applications such as image matching and map matching. Most subgraph matching methods fail to perform node-to-node matching under presence of such corruptions. We propose a method for identifying the node correspondence between a subgraph and a full graph in the inexact case without node labels in two steps - (a) extract the minimal unique topology preserving subset from the subgraph and find its feasible matching in the full graph, and (b) implement a consensus-based algorithm to expand the matched node set by pairing unique paths based on boundary commutativity. To demonstrate the effectiveness of the proposed method, a simulation is performed on the Erdos-Renyi random graphs and two case studies are performed on the image-based affine covariant features dataset and KITTI stereo dataset respectively. Going beyond the existing subgraph matching approaches, the proposed method is shown to have realistically sub-linear computational efficiency, robustness to random measurement noise, and good statistical properties. Our method is also readily applicable to the exact matching case without loss of generality.

A review of lane change estimation algorithms shows they typically rely on trajectory data with high-definition maps or camera-based data, both of which are costly. Departing from these approaches, our work introduces an innovative method that extracts lane changes solely from highway GPS data, eliminating the need for lane-level maps or camera input. We treat each trajectory as a combination of road curvatures which correspond to large scale variations, and lane changes and lane deviations which correspond to small scale variations. We project each trajectory into linear basis functions at different window sizes, aggregating the deviations for each point on the trajectory. We refer to this transformed space as basis-aligned coordinate system (BACS). Leveraging these BACS features, we propose two methods for extracting lane changes: (a) using an inverse operation on the transformed curves along with an optimization procedure to detect deviations, and (b) applying consistency measures across window sizes to identify anomalous regions as lane changes. We show that our approach can be theoretically justified and can robustly recover lane changes while simultaneously controlling for random noise. We utilize real trajectories collected in the Safety Pilot Model Deployment project and show the performance of our proposed approach through precision-recall curves when compared to two different ground truths based on maps and camera data respectively. We also provide ablation studies regarding the choice of window sizes, highlighting how the projection distances become smoother as a function of window size.

Pedestrian safety has become a critical issue due to the increase in pedestrian crashes every year, while proactive traffic safety management based on surrogate safety measures (SSMs) has been considered one of the key approaches to improving pedestrian safety. However, existing SSMs are developed based on the assumption that road users will maintain constant speed and direction. Risk estimations based on this assumption are less stable and more likely to be exaggerated. Considering the limitations of existing SSMs, this study has proposed a probabilistic framework for estimating the risk of pedestrian-vehicle conflicts at intersections. The proposed framework works by predicting the trajectories of vehicles and pedestrians using Gaussian process regression models and incorporating these results with the probability of vehicles making different maneuvers. The proposed framework has been evaluated using both simulated and real-world data collected at an intersection. The simulation results validated an increased estimated risk given time-critical pedestrian-vehicle conflicts, as well as a higher probability of the vehicle maneuver that led to such conflicts. This observation remained even when multiple conflicts arose from different directions. Moreover, experimental results using real-world data suggested that the proposed framework outperformed traditional time-to-collision (TTC) in terms of conflict prediction, quantification, and localization. For example, the proposed framework had a sensitivity of 0.92 in terms of conflict prediction, while TTC had a sensitivity of 0.62. Furthermore, the proposed framework required much less computation time compared to deep learning methods, which made it an optimal choice for proactive pedestrian safety solutions at intersections.

Current autonomous vehicle (AV) simulators are built to provide large-scale testing required to prove capabilities under varied conditions in controlled, repeatable fashion. However, they have certain failings including the need for user expertise and complex inconvenient tutorials for customized scenario creation. Simulation of Urban Mobility (SUMO) simulator, which has been presented as an open-source traffic simulation platform, has found use as an AV simulator but suffers from similar issues which makes it difficult for entry-level practitioners to utilize the simulator without significant time investment. In that regard, we provide two enhancements to SUMO simulator geared towards massively improving user experience and providing real-life like variability for surrounding traffic. Firstly, we calibrate a car-following model, Intelligent Driver Model (IDM), for highway and urban naturalistic driving data and sample randomly from the parameter distributions to create realistic background vehicle driving behavior. Secondly, we combine SUMO with OpenAI gym, creating a Python package placed in a docker container which can run simulations based on real world highway and urban layouts with generic output observations and input actions that can be processed via any AV pipeline. For the calibration, we provide results using simulated and real-life data. For the Sumo-Gym package, we showcase a simple AV platform which runs IDM and lane change throughout the highway loop and provide some qualitative results. Our aim through these enhancements is to provide an easy-to-use simulation environment which can be installed in any operating platform and can be readily used for AV testing and validation.

A common approach for defining a reward function for multi-objective reinforcement learning (MORL) problems is the weighted sum of the multiple objectives. The weights are then treated as design parameters dependent on the expertise (and preference) of the person performing the learning, with the typical result that a new solution is required for any change in these settings. This paper investigates the relationship between the reward function and the optimal value function for MORL; specifically addressing the question of how to approximate the optimal value function well beyond the set of weights for which the optimization problem was actually solved, thereby avoiding the need to recompute for any particular choice. We prove that the value function transforms smoothly given a transformation of weights of the reward function (and thus a smooth interpolation in the policy space). A Gaussian process is used to obtain a smooth interpolation over the reward function weights of the optimal value function for three well-known examples: Gridworld, Objectworld and Pendulum. The results show that the interpolation can provide robust values for sample states and actions in both discrete and continuous domain problems. Significant advantages arise from utilizing this interpolation technique in the domain of autonomous vehicles: easy, instant adaptation of user preferences while driving and true randomization of obstacle vehicle behavior preferences during training.

Quantifying near-field displacements can help enable a better understanding of earthquake physics and hazards. To date, established remote sensing techniques have failed to recover subcentimeter-level near-field displacements at the scale and resolution required for shallow fault physical investigations. In this paper, methods are developed to rapidly extract planar parameters, using fast parallel approaches and an alternative registration approach is employed to automatically match the planes extracted from pairwise temporally spaced mobile laser scanning (MLS) and Airborne laser scanning (ALS) data sets along the Napa fault. The features extracted from two temporally spaced point clouds are then used to calculate rigid-body transformation parameters. The production of robust and accurate deformation maps requires the selection of appropriate planar feature extraction and feature mapping criteria. Rigorously propagated point accuracy estimates are employed to produce realistic estimated errors for the transformation parameters. Displacements of each aggregate study area are computed separately from left and right sides of the fault and compared to be within 3 mm of alinement array displacements. Local differential displacements show distinct patterns which, compared to alinement array measurements, were found to agree within the confidence bounds. The findings demonstrate the ability to accurately estimate near-field deformations from repeated MLS or ALS scans of earthquake-prone urban areas. ALS is also used in conjunction with the MLS data sets, illustrating the algorithm's ability to accommodate different LiDAR collection modalities at subcentimeter-level accuracy. The automated planar extraction and registration is an important contribution to the study of near-field earthquake dynamics and can be used as input observations for future geological inversion models.