Discovering Closed-Loop Failures of Vision-Based Controllers via Reachability Analysis

arXiv:2211.02736v4 Announce Type: replace-cross
Abstract: Machine learning driven image-based controllers allow robotic systems to take intelligent actions based on the visual feedback from their environment. Understanding when these controllers might lead to system safety violations is important for their integration in safety-critical applications and engineering corrective safety measures for the system. Existing methods leverage simulation-based testing (or falsification) to find the failures of vision-based controllers, i.e., the visual inputs that lead to closed-loop safety violations. However, these techniques do not …

abstract analysis applications arxiv cs.ai cs.cv cs.lg cs.ro cs.sy eess.sy engineering environment feedback image integration intelligent loop machine machine learning robotic safety safety-critical safety measures systems type understanding via vision visual