Presentation
Software-Based Fault-Detection Technique for Object Tracking in Autonomous Vehicles
DescriptionAutonomous vehicles are gaining popularity and finding application in increasingly complex and strategic contexts. In this domain, Obstacle Detection and Avoidance Systems are crucial and must employ fault-detection and management techniques to maintain correct behavior. Aiming at enhancing fault-detection while producing little impact on system programmability, this paper introduces a novel and general monitoring technique, based on a user-directed observer design pattern, which aims at monitoring the validity of predicates over state variables of running algorithms. Results are evaluated on a real-world use-case from railway domain, showing how the proposed fault-detection mechanism can increase the overall reliability of the system.
Event Type
Work-in-Progress Poster
TimeWednesday, July 12th6:00pm - 7:00pm PDT
LocationLevel 2 Lobby
AI
Autonomous Systems
Cloud
Design
EDA
Embedded Systems
RISC-V
Security