I am Hansol Yoon, a PhD student in Professor Sriram Sankaranarayanan’s group and the Computer Science department at the University of Colorado Boulder. Also, I am a Republic of Korea Air Force Major, and I worked at the Avionics Software Development Center in the Korean Air Force before coming to the US.
My research focuses on runtime verification with applications in autonomous systems.
MS, Computer Science, University of Colorado Boulder, 2019
BS, Computer Science, Republic of Korea Air Force Academy, 2010
Semantic Anomaly Detection through Property Templates
We investigate the problem of detecting anomalous behaviors of deployed autonomous vehicles that are not consistent with “typical” or “baseline” behaviors. Detecting anomalies requires us to process large volumes of high dimensional data from on-board sensors including GPS, camera, LiDAR and RADAR. In this project, we propose a simple methodology of runtime monitoring through “property templates” that allow us to detect meaningful anomalies of interest to stakeholders, while at the same time reducing from a very high dimensional space to a lower dimensional space through the property templates. Next, we use Pareto frontiers to define anomalous instances in the data with respect to the property.
See more: Project page
Predictive Runtime Monitoring for Linear Stochastic Systems and Applications to Geofence Enforcement for UAVs.
Hansol Yoon, Yi Chou, Xin Chen, Eric Frew, and Sriram Sankaranarayanan,
International Conference on Runtime Verification (RV), 2019.