An agent model for fault-tolerant systems

This paper describes the use of fault tolerance in a multiagent system. Such an approach is based on the modeling of autonomous agents with planning capabilities. These capabilities are used by the agent to recover from faults occurring in its surrounding environment, e.g. hardware faults, or in its internal representation thereof, e.g. software faults. The expected fault-tolerant behavior is tested using fault injection either in the system described by the agent or in the environment in which the agent (system) is embedded into. Categories and Subject Descriptors I.2.11 [Artificial Intelligence]: Distributed Artificial Intelligence—Intelligent agents; D.4.5 [Software]: Operating Systems—Fault-tolerance General Terms Reliability, Algorithms, Languages Keywords CA actions, BDI Model, Planning, Declarative Goals

• Type: Conference
• Year: 2005
• Where: SAC
• Authors: Avelino F. Zorzo, Felipe Rech Meneguzzi
• Keywords: Applied Computing, Fault Tolerance, Hardware Faults, SAC 2005, Software Faults