COLLABORATIVE MULTI-AGENT VEHICLE FAULT DIAGNOSTIC SYSTEM & ASSOCIATED METHODOLOGY

• US 20120277949A1
• Filed: 04/29/2011
• Published: 11/01/2012
• Est. Priority Date: 04/29/2011
• Status: Active Grant

• Current Assignee: Toyota Motor Corporation
• Original Assignee: Toyota Motor Engineering & Manufacturing North America Incorporated (Toyota Motor Corporation)
• Inventors: GHIMIRE, Sandesh, PROKHOROV, Danil
• Granted Patent: US 8,543,280 B2
• Time in Patent Office: Days
• US Class Current: 701/31.4
• CPC Class Codes: G07C 5/008 communicating information to a remotely located station transmission systems for measured values G08C
• CPC Class Codes: G07C 5/085 using electronic data carriers

Abstract

A control unit executes a first agent of a first subsystem of a vehicle, the first agent including a fault detection agent and a communication agent. The fault detection agent monitors performance data of the first subsystem to identify a problem in the first subsystem, stores first candidate fault diagnosis processes corresponding to the first subsystem in a first memory, and searches for a fault diagnosis process which identifies a fault causing the problem in the first subsystem from among the first candidate fault diagnosis processes. The communication agent collaborates with a second agent when a search by the fault detection agent fails to find the fault diagnosis process which identifies the fault. The collaboration includes searching a second memory which stores second candidate fault diagnosis processes, sending the monitored performance data to the second agent, and exchanging search result data between the first and second agents.

Claims

1. A control unit comprising:

a processor configured to execute a first agent of a first subsystem of a vehicle, the first agent including a fault detection agent and a communication agent,

the fault detection agent configured to monitor performance data of the first subsystem to identify a problem in the first subsystem, to store first candidate fault diagnosis processes corresponding to the first subsystem in a first memory, and to search for a fault diagnosis process which identifies a fault causing the problem in the first subsystem from among the first candidate fault diagnosis processes, and

the communication agent configured to collaborate with a second agent when a search by the fault detection agent fails to find the fault diagnosis process which identifies the fault, the collaboration with the second agent including searching a second memory which stores second candidate fault diagnosis processes, sending the monitored performance data to the second agent, and exchanging search result data between the first and second agents.

2. The control unit according to claim 1, further comprising:

a network interface controller configured to establish a communication link between the first agent and a server agent when the collaboration with the second agent fails to result in finding the fault diagnosis process which identifies the fault,

the communication agent configured to receive a server candidate fault diagnosis process from the server agent.

3. The control unit according to claim 2, wherein the communication agent is configured to distribute the server candidate fault diagnosis process to the fault detection agent of the first agent and the second agent.

4. The control unit according to claim 3, wherein

the communication agent receives a result of the second agent applying the server candidate fault diagnosis process, and

the fault detection agent compares a result of the fault detection agent applying the server candidate fault diagnosis process to the result of the second agent applying the server candidate fault diagnosis process to identify the fault.

5. The control unit according to claim 4, wherein an event log of the identification of the fault is stored in the first memory.

6. The control unit according to claim 2, wherein the server agent is executed by a server operated remotely from the control unit.

7. The control unit according to claim 2, wherein

the network interface controller is configured to establish a communication link between the first agent, the second agent and the server agent by a network, and

the communication agent is configured to receive the server candidate fault diagnosis process from the server agent in response to:

(1) the server agent connecting to the network, and (2) the communication agent, in response to the server agent connecting to the network, transmitting the monitored performance data and the search result data exchanged between the first and second agents to the server agent.

8. The control unit according to claim 7, wherein the communication agent is further configured to transmit statistical and probability models of the monitored performance data to the server agent.

9. The control unit according to claim 1, wherein

the second agent corresponds to a second subsystem of the vehicle, and

the second agent is configured to monitor performance data of the second subsystem.

10. The control unit according to claim 9, wherein

the first agent is configured to identify a correlation between the monitored performance data of the first and second subsystems, and

the first agent searches the second memory for the fault diagnosis process which identifies the fault based on the correlation.

11. The control unit according to claim 9, further comprising:

a memory module including the first and second memories as discrete memory banks, wherein the processor is further configured to execute the second agent.

12. The control unit according to claim 1, wherein the first agent further includes a clustering agent configured to generate a statistical model and a probability model from the monitored performance data to predict a probability of a future problem.

13. The control unit according to claim 12, wherein the fault detection agent is configured to compare the monitored performance data of the first subsystem with data from the statistical and probability models of the clustering agent to identify and predict the problem in the first subsystem.

14. The control unit according to claim 1, wherein the search result data includes a result of applying at least a portion of a candidate fault diagnosis process to at least a portion of the monitored performance data.

15. The control unit according to claim 14, wherein the candidate fault diagnosis process includes at least one of (1) reading a selected portion of the monitored subsystem performance data, (2) comparing the monitored subsystem performance data to thresholds, (3) checking a timing requirement of the monitored subsystem performance data, and (4) reporting or logging the monitored subsystem performance data.

16. The control unit according to claim 1, wherein the first agent further includes a learning agent trained on healthy performance data and configured to identify the problem by comparing the monitored performance data to the healthy performance data.

17. The control unit according to claim 1, further comprising:

a network interface controller configured to communicatively link the first agent with the second agent, wherein the second agent is executed by a processor of another control unit.

18. A vehicle including a control unit according to claim 1.

19. A computer-readable medium storing a program for diagnosing vehicle faults, that, when executed by a computer, performs a method for identifying a fault causing a problem in a vehicle subsystem, the method comprising:

executing a first agent of a first subsystem of a vehicle, the first agent including a fault detection agent and a communication agent;

monitoring, by the fault detection agent, performance data of the first subsystem to identify a problem in the first subsystem;

searching, by the fault detection agent, for a fault diagnosis process which identifies a fault causing the problem in the first subsystem from among first candidate fault diagnosis processes stored in a first memory corresponding to the first subsystem; and

collaborating, by the communication agent, with a second agent when the searching by the fault detection agent fails to find the fault diagnosis process which identifies the fault, the collaborating including searching a second memory which stores second candidate fault diagnosis processes, sending the monitored performance data to the second agent, and exchanging search result data between the first and second agents.

20. A method for identifying a fault causing a problem in a vehicle subsystem, the method comprising:

executing a first agent of a first subsystem of a vehicle, the first agent including a fault detection agent and a communication agent;

monitoring, by the fault detection agent, performance data of the first subsystem to identify a problem in the first subsystem;

searching, by the fault detection agent, for a fault diagnosis process which identifies a fault causing the problem in the first subsystem from among first candidate fault diagnosis processes stored in a first memory corresponding to the first subsystem; and

collaborating, by the communication agent, with a second agent when the searching by the fault detection agent fails to find the fault diagnosis process which identifies the fault, the collaborating including searching a second memory which stores second candidate fault diagnosis processes, sending the monitored performance data to the second agent, and exchanging search result data between the first and second agents.