Abstract: Syfte och mål: The goal is to deliver:
- a planning algorithm for the integrated task of remote diagnostics and workshop diagnostics.
- a methodology for modeling the systems and distributed functions of the vehicle using Bayesian networks.
- a methodology for adding driver and mechanic actions and observations to a Bayesian network model.
- a prototype platform, a model of a selected system, and an integration of the model, a diagnostic inference engine, and the planner algorithm.
Förväntade effekter och resultat:
It will become necessary to have a computerized troubleshooting support system with guided diagnostics to help the mechanic. The current project will investigate how such guided diagnostics should be built for heavy-vehicle workshops. The troubleshooting problem is not isolated to the workshop. Instead we need to consider the whole chain, ranging from that a fault is detected on-board the vehicle, to that the vehicle is repaired and leaves the workshop.
Planerat upplägg och genomförande:
The project contains a research part and a prototyping part. The research part is mainly a continuation of ongoing PhD-projects for two PhD students in collaboration with Linköping University. The prototyping part aims to apply all of the project results to a real system and build a demonstration system with the purpose of being a major input to an industrialization of the results. Abstract: It will become necessary to have a computerized troubleshooting support system with guided diagnostics to help the mechanic. The current project will investigate how such guided diagnostics should be built for heavy-vehicle workshops. The troubleshooting problem is not isolated to the workshop. Instead we need to consider the whole chain, ranging from that a fault is detected on-board the vehicle, to that the vehicle is repaired and leaves the workshop.