The increase use of autonomous vehicles and other cyber-physical systems has been motivating the adoption of Runtime Verification (RV) techniques for embedded system platforms. This stems from the criticality of such systems, which call for the assurance of correct operation, both on value and time domains. However, traditional RV techniques (mostly based on code instrumentation) may inevitably pose significant overheads – both in performance and timeliness – due to their inherent intrusiveness, which make them clearly unfit for critical systems.
This project aims at advancing the state of the art in RV techniques by researching innovative (self-)adaptive observation and runtime verification methods, supported in non-intrusive monitoring machinery. The negative effects of traditional techniques (ranging from function call interception to source code annotation with observation points) are avoided, thus making this novel approach relevant to virtually all (critical) cyber-physical systems.
FCT nº 3732 | CAMPUS FRANCE nº 37932TF