Spectrum-Based Fault Localisation for Multi-Agent Systems
نویسندگان
چکیده
Diagnosing unwanted behaviour in Multi-Agent Systems (MASs) is crucial to ascertain agents’ correct operation. However, generation of MAS models is both error-prone and time intense, as it exponentially increases with the number of agents and their interactions. In this paper, we propose a light-weight, automatic debugging-based technique, coined ESFL-MAS, which shortens the diagnostic process, while only relying on minimal information about the system. ESFL-MAS uses a heuristic that quantifies the suspiciousness of an agent to be faulty; therefore, different heuristics may have different impact on the diagnostic quality. Our experimental evaluation shows that 10 out of 42 heuristics yield the best diagnostic accuracy (96.26% on average).
منابع مشابه
Sensitivity Analysis of Spectrum-based Fault Localisation for Multi-Agent Systems
Diagnosing unwanted behaviour in Multi-Agent Systems (MASs) is a crucial task to assure the correct operation of a system. A light-weight technique inspired by the software-engineeringoriented techniques, the we have coined Extended Spectrum-based Fault Localisation for Multi-Agent Systems (ESFL-MAS) can be used to shorten the diagnose cycle by reducing the testing effort. As the technique reli...
متن کاملFault Detection and Isolation of Multi-Agent Systems via Complex Laplacian
This paper studies the problem of fault detection and isolation (FDI) for multi-agent systems (MAS) via complex Laplacian subject to actuator faults. A planar formation of point agents in the plane using simple and linear interaction rules related to complex Laplacian is achieved. The communication network is a directed, and yet connected graph with a fixed topology. The loss of symmetry in the...
متن کاملFault diagnosis in a distillation column using a support vector machine based classifier
Fault diagnosis has always been an essential aspect of control system design. This is necessary due to the growing demand for increased performance and safety of industrial systems is discussed. Support vector machine classifier is a new technique based on statistical learning theory and is designed to reduce structural bias. Support vector machine classification in many applications in v...
متن کاملRethinking Multi-Layer Multi-Domain Network Monitoring
Accurate and efficient network fault localisation based on monitoring is obviously one of the vital but also formidable tasks for successful network operations. With proliferations of large-scale network services, e.g. Géant E2E links, monitoring and fault localisation across multiple domains with data obtained from different network layers are becoming unprecedentedly important. Many efforts h...
متن کاملModel-based Approach for Multi-sensor Fault Identification in Power Plant Gas Turbines
In this paper, the multi-sensor fault diagnosis in the exhaust temperature sensors of a V94.2 heavy duty gas turbine is presented. A Laguerre network-based fuzzy modeling approach is presented to predict the output temperature of the gas turbine for sensor fault diagnosis. Due to the nonlinear dynamics of the gas turbine, in these models the Laguerre filter parts are related to the linear d...
متن کامل