Distributed Fault Diagnosis for Networked Embedded Systems

In modern automotive vehicles, several Electronic Control Units (ECUs) communicate over a local network. Each ECU is connected to a number of components, e.g. sensors and actuators that are monitored by a diagnostic system to make sure that the components are operating correctly. The diagnostic system usually consists of a number of precompiled tests, simple or complex, to perform the monitoring.When a component becomes faulty, all tests involving that specific component should become invalidated and the diagnostic system should assign a Diagnostic Trouble Code (DTC) to each component that could possibly be faulty…


1 Introduction
1.1 Background
1.2 Objective
1.3 Approach
1.4 Contribution
1.5 Delimitations and Assumptions
1.6 Target Group
1.7 Related Work
2 Model Based Diagnosis
2.1 Introduction to Model Based Diagnosis
2.2 Artificial Intelligence and Fault Diagnosis
2.2.1 Behavioral Modes
2.2.2 Diagnoses
2.2.3 Conflicts
2.2.4 Relations between Diagnoses and Conflicts
2.2.5 Diagnostic Tests
2.3 Local Algorithms
2.3.1 Reiter’s Algorithm
2.3.2 Isolation with Generalized Fault Modes
2.3.3 Virtual Components
3 Distributed Systems
3.1 Properties of Distributed Systems
3.1.1 Transparency
3.1.2 Openness
3.1.3 Scalability
3.2 Hardware Concepts
3.2.1 The CAN Bus
vii4 Distributed Diagnostic Systems
4.1 The Network Architecture
4.2 Current Diagnostic System
4.2.1 The Goal with the Distributed Diagnostic System
4.3 Components, Signals and Objects
4.4 Signals – Inputs and Outputs
4.5 Local and Global Diagnosis
4.5.1 Two Ways of Calculating the Global Diagnosis
4.5.2 The Combinatorial Problem
4.5.3 Merging Minimal Cardinality Diagnoses
4.6 Centralized or Distributed Diagnosis
4.6.1 Centralized Diagnosis and Decentralized Diagnosis
4.6.2 Distributed Diagnosis
4.7 Sharing Diagnostic Information
4.7.1 Sharing Conflicts
4.7.2 Sharing Diagnoses
4.7.3 The Information to Share
4.7.4 Focusing on Probable Diagnosis
4.7.5 Problems with Component Representation
5 ProposedMethods for Distributed Diagnosis
5.1 Model for Distributed Diagnosis
5.2 Algorithms for Distributed Diagnosis
5.2.1 Method 1: Sharing Conflicts
5.2.2 Method 2: Sharing Diagnoses
5.2.3 Method 3: Sharing Diagnoses Extended
5.3 Discussion Concerning the Limitations and Assumptions
6 Implementation in an Embedded System
6.1 Hardware Setup
6.2 Software Description
6.3 Processes in Embedded Systems
6.4 Data Transferring on a CAN Bus
6.4.1 Protocol Design
6.4.2 Transparency
6.5 Memory Structure
6.5.1 Memory Conflicts
6.6 Time Handling
6.6.1 Diagnosis Executed in a Fixed Timed Loop
6.6.2 Diagnosis Executed in the Background Process
6.6.3 Synchronization
6.7 Using Reiter’s Algorithm in Distributed Diagnosis
6.8 Performance of the Implementation
7 Conclusions
8 FutureWork

Author: Hallgren, Dan,Skog, Håkan

Source: Linköping University

Download URL 2: Visit Now

Leave a Comment