Sensor Failure Detection System And Method

• US 20110153035A1
• Filed: 12/22/2009
• Published: 06/23/2011
• Est. Priority Date: 12/22/2009
• Status: Abandoned Application

• Current Assignee: Caterpillar Incorporated
• Original Assignee: Caterpillar Incorporated
• Inventors: Grichnik, Anthony J., Felty, Timothy J., Tippett, Tyler J., Mason, James R., Zobrist, Brandon P.
• Application Number:
  US12/644,497
• Publication Number:
  US 20110153035A1
• Time in Patent Office: Days
• US Class Current: 700/21
• CPC Class Codes: G05B 2219/2641 Fork lift, material handling vehicle
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks

Abstract

This disclosure relates to a system and method for detecting sensor failure and controlling a machine. The system includes a plurality of physical sensors, an electronic control module operably connected to the plurality of physical sensors, and a data storage unit operably connected to the electronic control module. The electronic control module is configured to retrieve calibration data associated with a plurality of input parameters, obtain a set of values of the plurality of input parameters, calculate a Mahalanobis distance of the set of values of the input parameters based on the calibration data, increment an evidence score of an input parameter if the Mahalanobis distance exceeds a threshold Mahalanobis distance value; and command a machine action when the evidence score of an input parameter exceeds an threshold evidence score value.

Claims

M1. M1. A method of controlling a machine, comprising:

retrieving calibration data associated with a plurality of input parameters,

wherein at least one of the plurality of input parameters represents data from a physical sensor on a machine

obtaining a set of values of the plurality of input parameters;

calculating a Mahalanobis distance of a set of values of the input parameters based on the calibration data;

incrementing an evidence score of an input parameter if the Mahalanobis distance exceeds a distance threshold; and

commanding a machine action when the evidence score of an input parameter exceeds an evidence score threshold.

M2. M2. The method of claim M1, including the step of:

calculating a standard deviation of at least one of the plurality of input parameters and incrementing the evidence score of an input parameter.

M3. M3. The method of claim M1, including the step of:

calculating whether at least one of the plurality of input parameters falls within a predetermined range of values.

M4. M4. The method of claim M1, wherein the machine action includes providing an indication that one of the plurality of physical sensors has failed.

M5. M5. The method of claim M4, wherein the indication includes an indication that the mode of sensor failure was a soft failure.

M6. M6. The method of claim M1, including the step of:

calculating a Mahalanobis distance of a subset the set of values of the input parameters based on the calibration data and incrementing the evidence score of an input parameter.

M7. M7. The method of claim M1, wherein the evidence score is an integer.

M8. M8. The method of claim M1, wherein the step of commanding a machine action includes causing a machine control system to use a virtual sensor in place of at least one of the plurality of physical sensors.

M9. M9. The method of claim M1, wherein the step of commanding a machine action includes removing power to the machine.

M10. M10. The method of claim M1, wherein the step of commanding a machine action includes deactivating a machine subsystem.

S1. S1. A system for detecting sensor failure on a machine, the system comprising:

a plurality of physical sensors;

an electronic control module operably connected to the plurality of physical sensors and a data storage unit, and configured to;

retrieve reference calibration data associated with a plurality of input parameters,

wherein at least one of the plurality of input parameters represents data from at least one of the plurality of physical sensors;

obtain a set of values of the plurality of input parameters;

calculate a Mahalanobis distance of the set of values of the input parameters to the reference calibration data;

increment an evidence score of an input parameter if the Mahalanobis distance exceeds a threshold distance value; and

command a machine action when the evidence score of an input parameter exceeds an threshold evidence score value.

S2. S2. The system of claim S1, wherein the electronic control module is further configured to calculate the standard deviation of at least one value of the plurality of physical sensors.

S3. S3. The system of claim S1, wherein the machine action includes causing a machine control system to use a virtual sensor in place of at least one of the plurality of physical sensors.

S4. S4. The system of claim S1, wherein the machine action includes providing an indication of a sensor failure.

S5. S5. The system of claim S1, wherein the machine action includes removing power to the machine.

S6. S6. The system of claim S1, wherein the machine action includes deactivating a machine subsystem.

S7. S7. The system of claim S1, wherein the electronic control module is further configured to send data relating to at least one of the plurality of physical sensors off-board the machine.

S8. S8. The system of claim S1, wherein the electronic control module is further configured to calculate a Mahalanobis distance of a subset of the set of values of the input parameters based on the calibration data.