2019/12/18 by Nauman Sohani, Sohani, Nauman, Geunseob Oh +3
Computer Science · Decision Sciences · Engineering · Mathematics · #Algorithm #Artificial intelligence #Boundary (topology) #Classifier (UML) #Computer science #Control (management) #Data Visualization and Analytics #Data mining #Data science #Engineering #FOS: Computer and information sciences #FOS: Electrical engineering #Formal Methods in Verification #Human behavior #Intersection (aeronautics) #Mathematics #Parametric statistics #Real world data #Robotics (cs.RO) #Simulation Techniques and Applications #Systems and Control (eess.SY) #Transport engineering #cs.RO #cs.SY #eess.SY #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1912.08361
published in arXiv (Cornell University) (Cornell University) · ACC 2020. The first two authors contributed equally to this work
openalex publication_date 2019/12/18 · arxiv created 2020/04/27 · arxiv updated 2020/04/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We propose a novel framework to differentiate between vehicle trajectories originating from human and non-human drivers by constructing a data-driven boundary using parametric signal temporal logic (STL). Such construction allows us to evaluate the trajectories, detect rare-events, and reduce the uncertainty of driver behaviors when it assumes the form of a disturbance in control synthesis and evaluation problems. We train a classifier that separates admissible (i.e. human) examples - which arise from real-world demonstrations - and inadmissible (i.e. non-human) examples that are generated by falsifying specifications synthesized from the same real-world driving data. Proceeding in this fashion allows for finding a reasonable boundary of human behaviors exhibited in real-world driving records. The framework is demonstrated using a case study involving a human-driven vehicle approaching a signalized intersection.