This paper presents crash rate benchmarks for evaluating US-based Automated Driving Systems (ADS) for multiple urban areas. The purpose of this study was to extend prior benchmarks focused only on surface streets to additionally capture freeway crash risk for future ADS safety performance assessments. Using publicly available police-reported crash and vehicle miles traveled (VMT) data, the methodology details the isolation of in-transport passenger vehicles, road type classification, and crash typology. Key findings revealed that freeway crash rates exhibit large geographic dependence variations with any-injury-reported crash rates being nearly 3.5 times higher in Atlanta (2.4 IPMM; the highest) when compared to Phoenix (0.7 IPMM; the lowest). The results show the critical need for location-specific benchmarks to avoid biased safety evaluations and provide insights into the vehicle miles traveled (VMT) required to achieve statistical significance for various safety impact levels. The distribution of crash types depended on the outcome severity level. Higher severity outcomes (e.g., fatal crashes) had a larger proportion of single-vehicle, vulnerable road users (VRU), and opposite-direction collisions compared to lower severity (police-reported) crashes. Given heterogeneity in crash types by severity, performance in low-severity scenarios may not be predictive of high-severity outcomes. These benchmarks are additionally used to quantify at the required mileage to show statistically significant deviations from human performance. This is the first paper to generate freeway-specific benchmarks for ADS evaluation and provides a foundational framework for future ADS benchmarking by evaluators and developers.

Urban and transportation research has long sought to uncover statistically meaningful relationships between key variables and societal outcomes such as road safety, to generate actionable insights that guide the planning, development, and renewal of urban and transportation systems. However, traditional workflows face several key challenges: (1) reliance on human experts to propose hypotheses, which is time-consuming and prone to confirmation bias; (2) limited interpretability, particularly in deep learning approaches; and (3) underutilization of unstructured data that can encode critical urban context. Given these limitations, we propose a Multimodal Large Language Model (MLLM)-based approach for interpretable hypothesis inference, enabling the automated generation, evaluation, and refinement of hypotheses concerning urban context and road safety outcomes. Our method leverages MLLMs to craft safety-relevant questions for street view images (SVIs), extract interpretable embeddings from their responses, and apply them in regression-based statistical models. UrbanX supports iterative hypothesis testing and refinement, guided by statistical evidence such as coefficient significance, thereby enabling rigorous scientific discovery of previously overlooked correlations between urban design and safety. Experimental evaluations on Manhattan street segments demonstrate that our approach outperforms pretrained deep learning models while offering full interpretability. Beyond road safety, UrbanX can serve as a general-purpose framework for urban scientific discovery, extracting structured insights from unstructured urban data across diverse socioeconomic and environmental outcomes. This approach enhances model trustworthiness for policy applications and establishes a scalable, statistically grounded pathway for interpretable knowledge discovery in urban and transportation studies.

Autonomous vehicles (AVs) have significantly advanced in real-world deployment in recent years, yet safety continues to be a critical barrier to widespread adoption. Traditional functional safety approaches, which primarily verify the reliability, robustness, and adequacy of AV hardware and software systems from a vehicle-centric perspective, do not sufficiently address the AV's broader interactions and behavioral impact on the surrounding traffic environment. To overcome this limitation, we propose a paradigm shift toward behavioral safety, a comprehensive approach focused on evaluating AV responses and interactions within traffic environment. To systematically assess behavioral safety, we introduce a third-party AV safety assessment framework comprising two complementary evaluation components: Driver Licensing Test and Driving Intelligence Test. The Driver Licensing Test evaluates AV's reactive behaviors under controlled scenarios, ensuring basic behavioral competency. In contrast, the Driving Intelligence Test assesses AV's interactive behaviors within naturalistic traffic conditions, quantifying the frequency of safety-critical events to deliver statistically meaningful safety metrics before large-scale deployment. We validated our proposed framework using \texttt{Autoware.Universe}, an open-source Level 4 AV, tested both in simulated environments and on the physical test track at the University of Michigan's Mcity Testing Facility. The results indicate that \texttt{Autoware.Universe} passed 6 out of 14 scenarios and exhibited a crash rate of 3.01e-3 crashes per mile, approximately 1,000 times higher than average human driver crash rate. During the tests, we also uncovered several unknown unsafe scenarios for \texttt{Autoware.Universe}. These findings underscore the necessity of behavioral safety evaluations for improving AV safety performance prior to widespread public deployment.

In this work, we specialize contributions from prior work on data-driven trajectory generation for a quadrotor system with motor saturation constraints. When motors saturate in quadrotor systems, there is an ``uncontrolled drift" of the vehicle that results in a crash. To tackle saturation, we apply a control decomposition and learn a tracking penalty from simulation data consisting of low, medium and high-cost reference trajectories. Our approach reduces crash rates by around $49\%$ compared to baselines on aggressive maneuvers in simulation. On the Crazyflie hardware platform, we demonstrate feasibility through experiments that lead to successful flights. Motivated by the growing interest in data-driven methods to quadrotor planning, we provide open-source lightweight code with an easy-to-use abstraction of hardware platforms.

SAE Level 4 Automated Driving Systems (ADSs) are deployed on public roads, including Waymo's Rider-Only (RO) ride-hailing service (without a driver behind the steering wheel). The objective of this study was to perform a retrospective safety assessment of Waymo's RO crash rate compared to human benchmarks, including disaggregated by crash type. Eleven crash type groups were identified from commonly relied upon crash typologies that are derived from human crash databases. Human benchmarks were aligned to the same vehicle types, road types, and locations as where the Waymo Driver operated. Waymo crashes were extracted from the NHTSA Standing General Order (SGO). RO mileage was provided by the company via a public website. Any-injury-reported, Airbag Deployment, and Suspected Serious Injury+ crash outcomes were examined because they represented previously established, safety-relevant benchmarks where statistical testing could be performed at the current mileage. Data was examined over 56.7 million RO miles through the end of January 2025, resulting in a statistically significant lower crashed vehicle rate for all crashes compared to the benchmarks in Any-Injury-Reported and Airbag Deployment, and Suspected Serious Injury+ crashes. Of the crash types, V2V Intersection crash events represented the largest total crash reduction, with a 96% reduction in Any-injury-reported (87%-99% CI) and a 91% reduction in Airbag Deployment (76%-98% CI) events. Cyclist, Motorcycle, Pedestrian, Secondary Crash, and Single Vehicle crashes were also statistically reduced for the Any-Injury-Reported outcome. There was no statistically significant disbenefit found in any of the 11 crash type groups. This study represents the first retrospective safety assessment of an RO ADS that made statistical conclusions about more serious crash outcomes and analyzed crash rates on a crash type basis.

Ensuring the safety of autonomous vehicles (AVs) requires identifying rare but critical failure cases that on-road testing alone cannot discover. High-fidelity simulations provide a scalable alternative, but automatically generating realistic and diverse traffic scenarios that can effectively stress test AV motion planners remains a key challenge. This paper introduces CRASH - Challenging Reinforcement-learning based Adversarial scenarios for Safety Hardening - an adversarial deep reinforcement learning framework to address this issue. First CRASH can control adversarial Non Player Character (NPC) agents in an AV simulator to automatically induce collisions with the Ego vehicle, falsifying its motion planner. We also propose a novel approach, that we term safety hardening, which iteratively refines the motion planner by simulating improvement scenarios against adversarial agents, leveraging the failure cases to strengthen the AV stack. CRASH is evaluated on a simplified two-lane highway scenario, demonstrating its ability to falsify both rule-based and learning-based planners with collision rates exceeding 90%. Additionally, safety hardening reduces the Ego vehicle's collision rate by 26%. While preliminary, these results highlight RL-based safety hardening as a promising approach for scenario-driven simulation testing for autonomous vehicles.

Deployed SAE level 4+ Automated Driving Systems (ADS) without a human driver are currently operational ride-hailing fleets on surface streets in the United States. This current use case and future applications of this technology will determine where and when the fleets operate, potentially resulting in a divergence from the distribution of driving of some human benchmark population within a given locality. Existing benchmarks for evaluating ADS performance have only done county-level geographical matching of the ADS and benchmark driving exposure in crash rates. This study presents a novel methodology for constructing dynamic human benchmarks that adjust for spatial and temporal variations in driving distribution between an ADS and the overall human driven fleet. Dynamic benchmarks were generated using human police-reported crash data, human vehicle miles traveled (VMT) data, and over 20 million miles of Waymo's rider-only (RO) operational data accumulated across three US counties. The spatial adjustment revealed significant differences across various severity levels in adjusted crash rates compared to unadjusted benchmarks with these differences ranging from 10% to 47% higher in San Francisco, 12% to 20% higher in Maricopa, and 7% lower to 34% higher in Los Angeles counties. The time-of-day adjustment in San Francisco, limited to this region due to data availability, resulted in adjusted crash rates 2% lower to 16% higher than unadjusted rates, depending on severity level. The findings underscore the importance of adjusting for spatial and temporal confounders in benchmarking analysis, which ultimately contributes to a more equitable benchmark for ADS performance evaluations.

Autonomous vehicles (AVs) face significant threats to their safe operation in complex traffic environments. Adversarial training has emerged as an effective method of enabling AVs to preemptively fortify their robustness against malicious attacks. Train an attacker using an adversarial policy, allowing the AV to learn robust driving through interaction with this attacker. However, adversarial policies in existing methodologies often get stuck in a loop of overexploiting established vulnerabilities, resulting in poor improvement for AVs. To overcome the limitations, we introduce a pioneering framework termed Vulnerability-aware and Curiosity-driven Adversarial Training (VCAT). Specifically, during the traffic vehicle attacker training phase, a surrogate network is employed to fit the value function of the AV victim, providing dense information about the victim's inherent vulnerabilities. Subsequently, random network distillation is used to characterize the novelty of the environment, constructing an intrinsic reward to guide the attacker in exploring unexplored territories. In the victim defense training phase, the AV is trained in critical scenarios in which the pretrained attacker is positioned around the victim to generate attack behaviors. Experimental results revealed that the training methodology provided by VCAT significantly improved the robust control capabilities of learning-based AVs, outperforming both conventional training modalities and alternative reinforcement learning counterparts, with a marked reduction in crash rates. The code is available at https://github.com/caixxuan/VCAT.

The public, regulators, and domain experts alike seek to understand the effect of deployed SAE level 4 automated driving system (ADS) technologies on safety. The recent expansion of ADS technology deployments is paving the way for early stage safety impact evaluations, whereby the observational data from both an ADS and a representative benchmark fleet are compared to quantify safety performance. In January 2024, a working group of experts across academia, insurance, and industry came together in Washington, DC to discuss the current and future challenges in performing such evaluations. A subset of this working group then met, virtually, on multiple occasions to produce this paper. This paper presents the RAVE (Retrospective Automated Vehicle Evaluation) checklist, a set of fifteen recommendations for performing and evaluating retrospective ADS performance comparisons. The recommendations are centered around the concepts of (1) quality and validity, (2) transparency, and (3) interpretation. Over time, it is anticipated there will be a large and varied body of work evaluating the observed performance of these ADS fleets. Establishing and promoting good scientific practices benefits the work of stakeholders, many of whom may not be subject matter experts. This working group's intentions are to: i) strengthen individual research studies and ii) make the at-large community more informed on how to evaluate this collective body of work.

With fully automated driving systems (ADS; SAE level 4) ride-hailing services expanding in the US, we are now approaching an inflection point, where the process of retrospectively evaluating ADS safety impact can start to yield statistically credible conclusions. An ADS safety impact measurement requires a comparison to a "benchmark" crash rate. This study aims to address, update, and extend the existing literature by leveraging police-reported crashes to generate human crash rates for multiple geographic areas with current ADS deployments. All of the data leveraged is publicly accessible, and the benchmark determination methodology is intended to be repeatable and transparent. Generating a benchmark that is comparable to ADS crash data is associated with certain challenges, including data selection, handling underreporting and reporting thresholds, identifying the population of drivers and vehicles to compare against, choosing an appropriate severity level to assess, and matching crash and mileage exposure data. Consequently, we identify essential steps when generating benchmarks, and present our analyses amongst a backdrop of existing ADS benchmark literature. One analysis presented is the usage of established underreporting correction methodology to publicly available human driver police-reported data to improve comparability to publicly available ADS crash data. We also identify important dependencies in controlling for geographic region, road type, and vehicle type, and show how failing to control for these features can bias results. This body of work aims to contribute to the ability of the community - researchers, regulators, industry, and experts - to reach consensus on how to estimate accurate benchmarks.