Systems and methods for predicting consequences of misinterpretation of user commands in automated systems

• US 7,437,297 B2
• Filed: 01/27/2005
• Issued: 10/14/2008
• Est. Priority Date: 01/27/2005
• Status: Active Grant

• Current Assignee: Hyundai Motor Company (Hyundai Motor Group), Kia Corp.
• Original Assignee: International Business Machines Corporation
• Inventors: Viswanathan, Mahesh, Kanevsky, Dimitri, Sicconi, Roberto, Chaar, Jarir Kamel
• Primary Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US11/044,316
• Publication Number:
  US 20060167696A1
• Time in Patent Office: 1,356 Days
• Field of Search: 704/275
• US Class Current: 704/275
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

Systems and methods are provided for processing and executing commands in automated systems. For example, command processing systems and methods are provided which can automatically determine, evaluate or otherwise predict consequences of execution of misrecognized or misinterpreted user commands in automated systems and thus prevent undesirable or dangerous consequences that can result from execution of misrecognized/misinterpreted commands.

Claims

1. A method for processing commands in an automated system, comprising:

receiving a user command;

processing the user command;

evaluating consequences of executing the user command based on command processing results, wherein evaluating consequences of executing the user command comprises determining a potential consequence of executing the user command, determining a likelihood that the potential consequence can actually occur, and classifying an impact of the potential consequence based on the determined likelihood that the potential consequence can actually occur; and

performing a preventative action if it is determined that execution of the user command will result in an undesirable consequence.

2. The method of claim 1, wherein processing the user command comprises interpreting the user command.

3. The method of claim 2, wherein processing further comprises determining a degree of ambiguity of interpretation of the user command.

4. The method of claim 3, further comprising confirming the accuracy of the user command as interpreted, if the degree of ambiguity of interpretation is not acceptable.

5. The method of claim 4, wherein confirming the accuracy of the user command comprises automatically re-interpreting the user command using another interpretation process.

6. The method of claim 4, wherein confirming the accuracy of the user command comprises querying the user for confirmation.

7. The method of claim 3, further comprising the step of executing the user command as interpreted if the degree of ambiguity of interpretation is acceptable.

8. The method of claim 1, wherein interpreting the user command comprises performing a user state recognition process to evaluate the user'"'"'s state.

9. The method of claim 1, wherein determining a likelihood comprises:

receiving information that indicates current environmental conditions;

determining a relation, if any, between the potential consequence and the current environmental conditions; and

determining a likelihood that the potential consequence can actually occur based on related current environmental conditions.

10. The method of claim 1, wherein processing the user command comprises determining a plurality of alternative interpretations of the user commands and a likelihood measure of each alternative interpretation.

11. The method of claim 1, wherein performing a preventative action comprises automatically generating an alert signal or notification message to provide notification of the undesirable consequence of executing the user command.

12. The method of claim 1, wherein performing a preventative action comprises generating a user query to obtain clarification of the user command.

13. The method of claim 12, wherein generating a user query comprises compiling a question to request user confirmation of the user command as interpreted or an alternative interpretation of the user command.

14. The method of claim 1, wherein performing a preventative action comprises preventing execution of the command if it is determined that execution of the user command will result in a consequence that does not meet a predetermined threshold of desirability.

15. The method of claim 1, wherein performing a preventative action comprises automatically re-processing the user command using one or more additional data processing modes.

16. The method of claim 1, wherein performing a preventative action comprises executing the command regardless of the undesirable consequence and providing an alert or notification of the undesirable consequence substantially contemporaneously with execution of the command.

17. The method of claim 1, wherein interpreting the user request comprises performing speech recognition to process a voice command.

18. The method of claim 1, wherein interpreting the user command comprises performing natural language understanding to process a voice command.