Processing sensor logs

• US 10,402,742 B2
• Filed: 12/12/2017
• Issued: 09/03/2019
• Est. Priority Date: 12/16/2016
• Status: Active Grant

• Current Assignee: Palantir Technologies Incorporated
• Original Assignee: Palantir Technologies Incorporated
• Inventors: Spiro, Ezra, Menck, Andre Frederico Cavalheiro, Maag, Peter, Powell, Thomas
• Primary Examiner(s): Islam, Mohammad K
• Application Number:
  US15/838,658
• Publication Number:
  US 20180173813A1
• Time in Patent Office: 630 Days
• CPC Class Codes: G01D 9/00 Recording measured values
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
• CPC Class Codes: G05B 23/0237 based on parallel systems, e.g. comparing signals produced at the same time by same type systems and detect faulty ones by noticing differences among their responses
• CPC Class Codes: G06F 16/901 Indexing; Data structures therefor; Storage structures for retrieval from the web G06F16/951
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
• CPC Class Codes: G06F 2218/16 by matching signal segments
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A method includes accessing a first sensor log and a corresponding first reference log. Each of the first sensor log and the first reference log includes a series of measured values of a parameter according to a first time series. The method also includes accessing a second sensor log and a corresponding second reference log. Each of the second sensor log and the second reference log includes a series of measured values of a parameter according to a second time series. The method also includes dynamically time warping the first reference log and/or second reference log by a first transformation between the first time series and a common time-frame and/or a second transformation between the second time series and the common time-frame. The method also includes generating first and second warped sensor logs by applying the or each transformation to the corresponding ones of the first and second sensor logs.

Claims

1. A computer-implemented method of processing sensor logs, wherein the method is performed using one or more processors or dedicated hardware, the method comprising:

accessing a first sensor log and a corresponding first reference log, each of the first sensor log and the first reference log comprising a first series of measured values of a first parameter and a second parameter according to a first time series, the first sensor log comprising measured values collected from a first sensor according to a regular schedule;

accessing a second sensor log and a corresponding second reference log, each of the second sensor log and the second reference log comprising a second series of measured values of the first parameter and the second parameter according to a second time series, the second sensor log comprising at least one controller command collected from any of the first sensor or a second sensor;

dynamically time warping the first reference log and the second reference log by a first transformation between the first time series and a common time-frame, and a second transformation between the second time series and the common time-frame, the dynamically time warping the first and second reference logs comprising;

dividing the first time series into a plurality of sub-periods in dependence upon the first reference log;

dividing the second time series into a plurality of sub-periods in dependence upon the second reference log;

for each of the first and second reference logs;

aggregating sub-periods of the same type;

arranging aggregated sub-periods in dependence upon sub-period type; and

setting each type of sub-period to correspond to an interval of the common time-frame;

generating first and second warped sensor logs by applying each of the first and second transformations to the corresponding first and second sensor logs, thereby aligning at least a first section of the first warped sensor log with at least one corresponding second section of the second warped sensor log, the generated first and second warped sensor logs facilitating a subsequent controlling of one or more machines based on the first and second warped sensor logs; and

causing a presentation of a user interface, the user interface including a representation of the at least a first section of the first warped sensor log and the at least one corresponding second section of the second warped sensor log.

2. The method of claim 1, wherein the first time series and the second time series are divided into an equal number of sub-periods.

3. The method of claim 1, wherein each sub-period has a type determined in dependence upon the corresponding first or second reference log.

4. The method of claim 1, wherein aggregating sub-periods of the same type comprises arranging the sub-periods of that type consecutively.

5. The method of claim 1, wherein aggregating sub-periods of the same type comprises:

setting each sub-period of that type to correspond to the same interval of the common time-frame; and

calculating, based on the sub-periods of the same type, a single value for each time within the interval of the common time-frame.

6. The method of claim 5, further comprising:

accessing a plurality of sensor logs and a corresponding plurality of reference logs, each sensor log and reference log comprising a series of measured values of a parameter according to a time series, each sensor log corresponding to a sensor type;

transforming the plurality of sensor logs to a single, common time-frame by dynamic time warping and

for each sensor type;

providing a baseline sensor log by determining and storing a series of averaged values according to a common time series spanning the common time-frame.

7. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

accessing a first sensor log and a corresponding first reference log, each of the first sensor log and the first reference log comprising a first series of measured values of a first parameter and a second parameter according to a first time series, the first sensor log comprising measured values collected from a first sensor according to a regular schedule;

accessing a second sensor log and a corresponding second reference log, each of the second sensor log and the second reference log comprising a second series of measured values of the first parameter and the second parameter according to a second time series, the second sensor log comprising at least one controller command collected from any of the first sensor or a second sensor;

dynamically time warping the first reference log and the second reference log by a first transformation between the first time series and a common time-frame, and a second transformation between the second time series and the common time-frame, the dynamically time warping the first and second reference logs comprising;

dividing the first time series into a plurality of sub-periods in dependence upon the first reference log;

dividing the second time series into a plurality of sub-periods in dependence upon the second reference log;

for each of the first and second reference logs;

aggregating sub-periods of the same type;

arranging aggregated sub-periods in dependence upon sub-period type; and

setting each type of sub-period to correspond to an interval of the common time-frame;

generating first and second warped sensor logs by applying each of the first and second transformations to the corresponding first and second sensor logs, thereby aligning at least a first section of the first warped sensor log with at least one corresponding second section of the second warped sensor log, the generated first and second warped sensor logs facilitating a subsequent controlling of one or more machines based on the first and second warped sensor logs; and

causing a presentation of a user interface, the user interface including a representation of the at least a first section of the first warped sensor log and the at least one corresponding second section of the second warped sensor log.

8. A system for processing sensor logs, the system comprising:

one or more processors;

memory storing instructions that, when executed by the one or more processors, cause the system to perform;

accessing a first sensor log and a corresponding first reference log, each of the first sensor log and the first reference log comprising a first series of measured values of a first parameter and a second parameter according to a first time series, the first sensor log comprising measured values collected from a first sensor according to a regular schedule;

accessing a second sensor log and a corresponding second reference log, each of the second sensor log and the second reference log comprising a second series of measured values of the first parameter and the second parameter according to a second time series, the second sensor log comprising at least one controller command collected from any of the first sensor or a second sensor;

dynamically time warping the first reference log and the second reference log by a first transformation between the first time series and a common time-frame, and a second transformation between the second time series and the common time-frame, the dynamically time warping the first and second reference logs comprising;

dividing the first time series into a plurality of sub-periods in dependence upon the first reference log;

dividing the second time series into a plurality of sub-periods in dependence upon the second reference log;

for each of the first and second reference logs;

aggregating sub-periods of the same type;

arranging aggregated sub-periods in dependence upon sub-period type; and

setting each type of sub-period to correspond to an interval of the common time-frame;

generating first and second warped sensor logs by applying each of the first and second transformations to the corresponding first and second sensor logs, thereby aligning at least a first section of the first warped sensor log with at least one corresponding second section of the second warped sensor log, the generated first and second warped sensor logs facilitating a subsequent controlling of one or more machines based on the first and second warped sensor logs; and

causing a presentation of a user interface, the user interface including a representation of the at least a first section of the first warped sensor log and the at least one corresponding second section of the second warped sensor log.

9. The system of claim 8, wherein the instructions further cause the system to divide the first and second time series into equal numbers of sub-periods.

10. The system of claim 8, wherein the instructions further cause the system is configured to determine a type of each sub-period in dependence upon the corresponding first or second reference log.

11. The system of claim 8, wherein the memory and one or more processors are implemented using:

one or more machines, each machine comprising one or more sensors, each sensor having a sensor type, wherein each machine is configured to output sensor logs comprising one or more series of measured values according to a time series and reference logs comprising one or more series of measured values according to a time series; and

wherein the instructions further cause the system to receive and process sensor logs and reference logs from the one or more machines.