About me
Comparing Narratives between Autonomous Vehicles Related Crash \and Uber Reported Crash through NLP Methods - As driverless automated driving systems (ADS) start to operate on public roads, there is an urgent need to understand how safely these systems are managing real-world traffic conditions. Much of the research focuses on estimating general crash patterns using categorical AV crash data, a comprehensive analysis of AV crash narratives remains limited. In this paper, the narratives are retrieved from Autonomous Vehicle Collision Reports from the California DMV and TNC Accidents and Incidents Reports from the CPUC. We study the latent variables of crashes, sentiments of the narratives, actors' involvement throughout the story, and interaction scenarios with vulnerable road users. These insights can further our understanding of AV safety.
