System and method for determination of machine state based on video and audio analytics

Patent metadata

Field	Value
Publication number	US-10324433-B2
Application number	US-201514676278-A
Country	US
Kind code	B2
Filing date	Apr 1, 2015
Priority date	Apr 1, 2015
Publication date	Jun 18, 2019
Grant date	Jun 18, 2019

Abstract

Systems and methods for analyzing and optimizing worksite operations based on video and or audio data are disclosed. One method includes receiving one or more models relating to a worksite, receiving first sensor data associated with the machine at the worksite, receiving second sensor data associated with an operation of the machine at the worksite, wherein the second sensor data is sourced from a sensor that is different from a sensor sourcing the first sensor data, determining, by the one or more processors, a machine state based at least on the first data and the second data, comparing the determined machine state to a modeled machine state represented by the received one or more models to classify site operations and/or detect an irregularity in site operations or an inefficiency in site operations, and generating a response based at least on the detected irregularity or inefficiency.

First claim

We claim:

1. A method for determining machine state, the method comprising:
receiving, by one or more processors, one or more predetermined models relating to a worksite or a machine at the worksite;
receiving, by the one or more processors, first sensor data associated with the machine at the worksite, the first sensor data comprising one or more of image data and audio data;
receiving, by the one or more processors, second sensor data associated with an operation of the machine at the worksite, wherein the second sensor data is sourced from a sensor that is different from a sensor sourcing the first sensor data;
determining, by the one or more processors, a machine state based at least on the first sensor data and the second sensor data, wherein the machine state includes at least one of a full load, empty load, payload material type, payload vs. air ratio, payload placement, payload compaction, payload water content, material amount moved, drop placement, excavator position, idle state, swing, dump position, deformation of machine, operator characteristic, and ground crew location relative to machine;
comparing the determined machine state to a modeled machine state represented by the received one or more predetermined models to detect an irregularity in site operations or an inefficiency in site operations; and
generating a response based at least on the detected irregularity or inefficiency and transmitting the response to one or more of an operator and a central station.

2. The method of claim 1 , wherein the one or more models comprises a worksite model and further comprising updating the worksite model based at least on the generated response.

3. The method of claim 1 , wherein the first sensor data is received via an onsite sensor and the second sensor data is received via an offsite sensor.

4. The method of claim 1 , wherein the first sensor data is received via an offsite sensor and the second sensor data is received via an onsite sensor.

5. The of claim 1 , wherein the determining the machine state comprises generating a signature representing the machine state.

6. The method of claim 1 , wherein the response comprises one or more of an audible indicator and a visual indicator.

7. A system for determining machine state, the system comprising:
a processor; and
a memory bearing instructions that, upon execution by the processor, cause the system at least to:
receive one or more predetermined models relating to a worksite or a machine at the worksite;
receive first sensor data associated with the machine at the worksite, the first sensor data comprising one or more of image data and audio data;
receive second sensor data associated with an operation of the machine at the worksite, wherein the second sensor data is sourced from a sensor that is different from a sensor sourcing the first sensor data;
determine a machine state based at least on the first sensor data and the second sensor data, wherein the machine state includes at least one of a full load, empty load, payload material type, payload vs. air ratio, payload placement, payload compaction, payload water content, material amount moved, drop placement, excavator position, idle state, swing, dump position, deformation of machine, operator characteristic, and ground crew location relative to machine; and
compare the determined machine state to a modeled machine state represented by the received one or more predetermined models to classify one or more site operations.

8. The system of claim 7 , wherein the one or more models comprises a worksite model and further comprising instructions that, upon execution by the processor, cause the system at least to update the worksite model based at least on the classified one or more site operations.

9. The system of claim 7 , wherein the first sensor data is received via an onsite sensor and the second sensor data is received via an offsite sensor.

10. The system of claim 7 , wherein the first sensor data is received via an offsite sensor and the second sensor data is received via an onsite sensor.

11. The of claim 7 , wherein the determining the machine state comprises generating a signature representing the machine state.

12. A non-transitory computer readable storage medium bearing instructions that, upon execution by one or more processors, effectuate operations for determining machine state, the non-transitory computer readable storage medium comprising:
receiving, by the one or more processors, first sensor data associated with a machine at a worksite, the first sensor data comprising one or more of image data and audio data;
receiving, by the one or more processors, second sensor data associated with an operation of the machine at the worksite, wherein the second sensor data is sourced from a sensor that is different from a sensor sourcing the first sensor data;
determining, by the one or more processors, a state signature based at least on the first sensor data and the second sensor data;
comparing the state signature to a stored state signature to determine a machine state, wherein the machine state includes at least one of a full load, empty load, payload material type, payload vs. air ratio, payload placement, payload compaction, payload water content, material amount moved, drop placement, excavator position, idle state, swing, dump position, deformation of machine, operator characteristic, and ground crew location relative to machine;
detecting an irregularity in site operations or an inefficiency in site operations based on the determined machine state; and
generating a response based at least on the detected irregularity or inefficiency.

13. The non-transitory computer readable storage medium of claim 12 , wherein the first sensor data is received via an onsite sensor and the second sensor data is received via an offsite sensor.

14. The non-transitory computer readable storage medium of claim 12 , wherein the first sensor data is received via an offsite sensor and the second sensor data is received via an onsite sensor.

15. The non-transitory computer readable storage medium of claim 12 , wherein the detecting an irregularity in site operations or an inefficiency in site operations comprises comparing the determined machine state to one or more models relating to the worksite.

16. The non-transitory computer readable storage medium of claim 12 , further comprising transmitting the response to one or more of an operator and a central station.

17. The method of claim 1 , wherein the determined machine state is a full load and the modeled machine state is a full load.

18. The method of claim 1 , wherein the determined machine state is an empty load and the modeled machine state is an empty load.

Assignees

• Caterpillar Inc

Inventors

• Sprock Christopher
• Baumann Ryan
• Talmaki Sanat
• Halepatali Praveen

Classifications

• G06T7/001 Primary

  using an image reference approach · CPC title

• G06T2207/30164

  Workpiece; Machine component · CPC title

• G05B19/00 Primary

  Program-control systems · CPC title