Safeware: system safety and computers
Safeware: system safety and computers
Formal methods: state of the art and future directions
ACM Computing Surveys (CSUR) - Special ACM 50th-anniversary issue: strategic directions in computing research
Real-Time Systems: Design Principles for Distributed Embedded Applications
Real-Time Systems: Design Principles for Distributed Embedded Applications
Hi-index | 0.00 |
In an effort to generate a safety assessment of various fault tolerant architectures for by-wire systems, we developed a methodology integrating system model checking including exercising all redundant aspects of the fault tolerant design. The system model definition includes both normal and failed modes of operation. The nominal behaviour operation and the failed mode of operation are checked via formal verification of some fault tolerance requirements. The comparison of the architectures can consist of several measures: fault recovery latency, fault tolerance to single point failures, fault tolerance to multiple point failures, and the complexity of the verification property observers. Failure coverage comparison of the designs can be facilitated by a proposed systematic management of verification of recovery properties for a set of failure modes.