Alexandre Bayen

Torsten Darrell, Mahyar Ghazanfari, Jordan Kam et al.

We investigate frameworks for post-flight safety analysis at non-towered airports using large language models (LLMs). Non-towered airports rely on the Common Traffic Advisory Frequency (CTAF) for air traffic coordination and experience frequent near mid-air collisions due to the pilot self-announcement communication protocol. We propose a general vision-language model (VLM) approach to analyze the transcribed CTAF radio communications in natural language, METeorological Aerodrome Report (METAR) weather data, Automatic Dependent Surveillance-Broadcast (ADS-B) flight trajectories, and Visual Flight Rules sectional charts of the airfield. We provide a preliminary study at Half Moon Bay Airport, with a qualitative real world case study and a quantitative evaluation using a new synthetic dataset of communications and weather modalities. We qualitatively evaluate our framework on real flight data using Gemini 2.5 Pro, demonstrating accurate identification of a right-of-way violation. The synthetic dataset is derived from real examples and includes a 12-category hazard taxonomy, and is used to benchmark three open-source (Qwen 2.5-7B, Mistral-7B, Gemma-2-9B) and three closed-source (GPT-4o, GPT-5.4, Claude Sonnet 4.6) LLM models on the subset of inputs related to CTAF and METAR. Even limited to CTAF and METAR inputs and open source LLMs, instances of our framework typically achieve a macro F1 score above 0.85 on a binary nominal/danger classification task. Future work includes a quantitative evaluation across all modalities and a larger number of real world examples. Taken together, our results suggest that VLM analysis of safety at non-towered airports may be a valuable future capability.

Jung Ho Park, Jordan Kam, Vishwanath Bulusu et al.

This article presents dynamic directional lane allocation in urban air mobility (UAM) corridors as a discrete-time mixed-integer linear program (MILP). This formulation activates, deactivates, and reverses lane direction as bi-directional airspace demand evolves. We model demand from disaggregate ground travel data by decomposing each trip into a multi-modal sequence with first-, middle-, and last-mile legs and routing the UAM-served middle-mile segment through a vertiport-side dispatch model. We use the San Francisco Bay Area as a case study by placing a multi-region spanning corridor between Contra Costa county and Silicon Valley. We find that the dynamic policy cuts unused airspace capacity by 5x, increases mean lane utilization from 36-48% to 67% at the same service level relative to baselines, and reduces commuting-population mean travel time by up to 21.6%. These results show that dynamic configuration of airspace capacity alleviates a significant percentage of the under-utilization issue of lane-based UAM airspace design and UAM concept of operations. This dynamic allocation also provides a safe, structural way to increase throughput, making UAM a more viable complement to multimodal door-to-door mobility systems.

Qing Lyu, Zhe Fu, Alexandre Bayen

Identifying safety-critical scenarios is essential for autonomous driving, but the rarity of such events makes supervised labeling impractical. Traditional rule-based metrics like Time-to-Collision are too simplistic to capture complex interaction risks, and existing methods lack a systematic way to verify whether statistical anomalies truly reflect physical danger. To address this gap, we propose an unsupervised anomaly detection framework based on a multi-agent Transformer that models normal driving and measures deviations through prediction residuals. A dual evaluation scheme has been proposed to assess both detection stability and physical alignment: Stability is measured using standard ranking metrics in which Kendall Rank Correlation Coefficient captures rank agreement and Jaccard index captures the consistency of the top-K selected items; Physical alignment is assessed through correlations with established Surrogate Safety Measures (SSM). Experiments on the NGSIM dataset demonstrate our framework's effectiveness: We show that the maximum residual aggregator achieves the highest physical alignment while maintaining stability. Furthermore, our framework identifies 388 unique anomalies missed by Time-to-Collision and statistical baselines, capturing subtle multi-agent risks like reactive braking under lateral drift. The detected anomalies are further clustered into four interpretable risk types, offering actionable insights for simulation and testing.