Personal protective equipment (PPE) with analytical stream processing for safety event detection

• US 9,998,804 B2
• Filed: 06/23/2017
• Issued: 06/12/2018
• Est. Priority Date: 06/23/2016
• Status: Active Grant

• Current Assignee: 3M Innovative Properties Company (3M Company)
• Original Assignee: 3M Innovative Properties Company (3M Company)
• Inventors: Awiszus, Steven T., Lobner, Eric C., Wurm, Michael G., Kanukurthy, Kiran S., Hu, Jia, Blackford, Matthew J., Mattson, Keith G., Jesme, Ronald D., Anderson, Nathan J.
• Primary Examiner(s): Tun, Nay
• Application Number:
  US15/631,870
• Publication Number:
  US 20170374436A1
• Time in Patent Office: 354 Days
• CPC Class Codes: A61F 9/06 Masks, shields or hoods for welders safety devices for welding in general F16P1/00
• CPC Class Codes: A62B 18/00 Breathing masks or helmets, e.g. affording protection against chemical agents or for use at high altitudes or incorporating a pump or compressor for reducing the inhalation effortA62B17/00 takes precedence; anaesthetic masks A61M16/06
• CPC Class Codes: A62B 27/00 Methods or devices for testing respiratory or breathing apparatus for high altitudesapparatus for testing gastightness in general G01M
• CPC Class Codes: A62B 9/00 Component parts for respiratory or breathing apparatus A62B19/00, A62B21/00, A62B23/00 take precedence
• CPC Class Codes: A62B 99/00 Subject matter not provided for in other groups of this subclass
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries
• CPC Class Codes: H04Q 9/00 Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom

Abstract

In some examples, a system includes an article of personal protective equipment (PPE) having at least one sensor configured to generate a stream of usage data; and an analytical stream processing component comprising: a communication component that receives the stream of usage data; a memory configured to store at least a portion of the stream of usage data and at least one model for detecting a safety event signature, wherein the at least one model is trained based as least in part on a set of usage data generated by one or more other articles of PPE of a same type as the article of PPE; and one or more computer processors configured to: detect the safety event signature in the stream of usage data based on processing the stream of usage data with the model, and generate an output in response to detecting the safety event signature.

Claims

1. A system comprising:

an article of personal protective equipment (PPE) having at least one sensor configured to generate a stream of usage data; and

an analytical stream processing component comprising;

a communication component that receives the stream of usage data from the at least one sensor of the article of PPE;

a memory configured to store at least a portion of the stream of usage data and at least one model for detecting a safety event signature indicative of a predicted occurrence of an imminent safety event, wherein the at least one model is trained based at least in part on a set of usage data generated, prior to receiving the stream of usage data, by one or more other articles of PPE of a same type as the article of PPE; and

one or more computer processors configured to;

detect the safety event signature indicative of the predicted occurrence of the imminent safety event in the stream of usage data based on processing the stream of usage data with the model, and

generate an output in response to detecting the safety event signature.

2. The system of claim 1, wherein the one or more computer processors:

select a training set comprising a set of training instances, each training instance comprising an association between usage data over a defined time duration and a safety event, wherein the usage data comprise one or more metrics that characterize at least one of a user, a work environment, or one or more articles of PPE; and

for each training instance in the training set, modify, based on particular usage data over the defined time duration and a particular safety event of the training instance, the model to change a likelihood predicted by the model for the particular safety event signature associated with the safety event in response to subsequent usage data over the defined time duration applied to the model.

3. The system of claim 2, wherein one or more training instances of the set of training instances are generated from use of one or more articles of PPE after the one or more computer processors detect the safety event signature.

4. The system of claim 1, wherein the safety event signature comprises at least one of an anomaly in a set of usage data, a pattern in a set of usage data, a particular set of occurrences of particular events over a defined period of time, a particular set of types of particular events over a defined period of time, a particular set of magnitudes of particular events over a defined period of time, a value that satisfies a threshold.

5. The system of claim 1, wherein the safety event signature is mapped to a safety event, wherein the safety event is associated with at least one of a worker, the article of PPE, an article of PPE other than the article of PPE, or a work environment.

6. The system of claim 5, wherein the safety event comprises at least one of an abnormal condition of worker behavior, an abnormal condition of the article of PPE, an abnormal condition in the work environment, or a violation of a safety regulation.

7. The system of claim 1, wherein the one or more computer processors:

prior to detection of the safety event signature, determine, based at least in part on the data stream of usage data, that the article of PPE is operating in a normal state; and

in response to detection of the safety event signature, determine that the article of PPE is not operating in the normal state.

8. The system of claim 7,

wherein the portion of the stream of usage data is a first portion of the stream of usage data,

wherein the safety event signature is a first safety event signature,

wherein the normal state corresponds to a second safety event signature indicative of a second predicted occurrence of a second imminent safety event,

wherein the first portion of the data stream corresponds to the first safety event signature, and

wherein a second portion of the data stream corresponds to the second safety event signature.

9. The system of claim 1, wherein at least one of the analytical stream processing component or the communication component is included in the article of PPE.

10. The system of claim 1, wherein at least one of the analytical stream processing component or the communication component is included in a worker device assigned a particular worker, wherein the article of PPE is assigned to the particular worker.

11. The system of claim 1, wherein at least one of the analytical stream processing component or the communication component is included in a computing device positioned at a location within a work environment in which a worker operates, wherein the article of PPE is assigned to the worker.

12. The system of claim 1, wherein the communication component is included in the article of PPE, a worker device assigned to a particular worker, or a computing device positioned at a location within a work environment, the system further comprising at least one server configured to receive the stream of usage data, store the at least one model, and detect the safety event signature in the stream of usage data based on processing the stream of usage data with the model.

13. The system of claim 12, further comprising a selection component configured to select at least one of the article of PPE, the worker device, the computing device, or the at least one server to detect the safety event signature in the stream of usage data.

14. The system of claim 13, wherein the selection component is incorporated in at least one of the article of PPE, the worker device, the computing device, or the at least one server to detect the safety event signature in the stream of usage data.

15. The system of claim 13, wherein the selection component is configured to select based on a power consumption associated with detecting the safety event signature, a latency associated with detecting the safety event signature, a connectivity status of the article of PPE, the worker device, the computing device, or the at least one server, a data type of the usage data, a data volume of the usage data, and the content of the usage data.

16. The system of claim 1, further comprising:

at least one sensor that generates usage data that characterizes at least a worker associated with the article of PPE or a work environment; and

wherein, to detect the safety event signature in the stream of usage, the one or more computer processors process the usage data that characterizes the worker associated with the article of PPE or the work environment.

17. The system of claim 1, wherein to generate the output in response to detecting the safety event signature, the one or more computer processors send a notification to at least one of the article of PPE, a hub associated with a user and configured to communicate with the article of PPE and at least one remote computing device, or a computing device associated with person who is not the user.

18. The system of claim 1, wherein to generate the output in response to detecting the safety event signature, the one or more computer processors send a notification that alters an operation of the article of PPE.

19. The system of claim 1, wherein to generate the output in response to detecting the safety event signature, the one or more computer processors output for display a user interface that indicates the safety event in association with at least one of a user, work environment, or the article of PPE.

20. The system of claim 1, wherein the article of PPE comprises at least one of an air respirator system, a fall protection device, a hearing protector, a head protector, a garment, a face protector, an eye protector, a welding mask, or an exosuit.

21. The system of claim 1,

wherein the article of PPE is included in a set of articles of PPE associated with a user,

wherein each article of PPE in the set of articles of PPE includes a motion sensor,

wherein the one or more computer processors:

receive a respective stream of usage data from each respective motion sensor of each respective article of PPE of the set of articles of PPE; and

to detect the safety event signature, the one or more computer processors detect the safety event signature corresponding to a relative motion that is based at least in part on the respective stream of usage data from each respective motion sensor.

22. The system of claim 1, wherein the stream of usage data comprises events, wherein each respective event is generated at a same defined interval, wherein each respective event includes a respective set of values that correspond to a same set of defined metrics, and wherein respective sets of values in different respective events are different.

23. The system of claim 22, wherein the set of defined metrics comprises one or more of a timestamp, characteristics of the article of PPE, characteristics of a worker associated with the article of PPE, or characteristics a work environment.

24. The system of claim 1, wherein at least one safety rule is mapped to at least one safety event, wherein the at least one safety event is mapped to the safety event signature, and wherein the safety event signature corresponds to at least the portion of the stream of usage data.

25. The system of claim 1, wherein to detect the safety event signature in the stream of usage data based on processing the stream of usage data with the model the one or more computer processors determine a set of one or more likelihoods associated with one or more safety event signatures, wherein the safety event signature is associated with a likelihood in the set of one or more likelihoods associated with one or more safety event signatures.

26. The system of claim 1, wherein to generate an output in response to detecting the safety event signature, the one or more processors generate a user interface that is based at least in part on a safety event that corresponds to the safety event signature.

27. The system of claim 26, wherein the user interface includes at least one input control that requires a responsive user input within a threshold time period, wherein the one or more computer processors:

in response to the threshold time period expiring without the responsive user input, perform at least one operation based at least in part on the threshold time period expiring without the responsive user input.

28. The system of claim 1, wherein the safety event signature corresponds to a safety event that indicates ergonomic stress that satisfies a threshold.

29. A computing device comprising:

one or more computer processors that receive a stream of usage data from at least one sensor of an article of personal protective equipment (PPE), wherein the article of PPE has the at least one sensor configured to generate the stream of usage data;

a memory that stores at least a portion of the stream of usage data and at least one model for detecting a safety event signature indicative of a predicted occurrence of an imminent safety event, wherein the at least one model is trained based at least in part on a set of usage data generated, prior to receiving the stream of usage data, by one or more other articles of PPE of a same type as the article of PPE;

wherein the one or more computer processors detect the safety event signature indicative of the predicted occurrence of the imminent safety event in the stream of usage data based on processing the stream of usage data with the model; and

wherein the one or more computer processors generate an output in response to detecting the safety event signature.

30. The computing device of claim 29, wherein the one or more computer processors:

select a training set comprising a set of training instances, each training instance comprising an association between usage data over a defined time duration and a safety event, wherein the usage data comprise one or more metrics that characterize at least one of a user, a work environment, or one or more articles of PPE; and

for each training instance in the training set, modify, based on particular usage data over the defined time duration and a particular safety event of the training instance, the model to change a likelihood predicted by the model for the particular safety event signature associated with the safety event in response to subsequent usage data over the defined time duration applied to the model.

31. The computing device of claim 30, wherein one or more training instances of the set of training instances are generated from use of one or more articles of PPE after the one or more computer processors detect the safety event signature.

32. The computing device of claim 29, wherein the safety event signature comprises at least one of an anomaly in a set of usage data, a pattern in a set of usage data, a particular set of occurrences of particular events over a defined period of time, a particular set of types of particular events over a defined period of time, a particular set of magnitudes of particular events over a defined period of time, a value that satisfies a threshold.

33. The computing device of claim 29, wherein the safety event signature is mapped to a safety event, wherein the safety event is associated with at least one of a worker, the article of PPE, an article of PPE other than the article of PPE, or a work environment.

34. The computing device of claim 33, wherein the safety event comprises at least one of an abnormal condition of worker behavior, an abnormal condition of the article of PPE, an abnormal condition in the work environment, or a violation of a safety regulation.

35. The computing device of claim 29, wherein the one or more computer processors:

prior to detecting the safety event signature, determine, based at least in part on the data stream of usage data, that the article of PPE is operating in a normal state; and

in response to detecting the safety event signature, determining that the article of PPE is not operating in the normal state.

36. The computing device of claim 35,

wherein the portion of the stream of usage data is a first portion of the stream of usage data,

wherein the safety event signature is a first safety event signature,

wherein the normal state corresponds to a second safety event signature indicative of a second predicted occurrence of a second imminent safety event,

wherein the first portion of the data stream corresponds to the first safety event signature, and

wherein a second portion of the data stream corresponds to the second safety event signature.

37. The computing device of claim 29, wherein the computing device is included at the article of PPE.

38. The computing device of claim 29, wherein computing device is included at a worker device assigned a particular worker, wherein the article of PPE is assigned to the particular worker.

39. The computing device of claim 29, wherein the computing device is positioned at a location within a work environment in which a worker operates, wherein the article of PPE is assigned to the worker.

40. The computing device of claim 29, wherein a communication component is communicatively coupled to the computing device, wherein the communication component is included in the article of PPE, a worker device assigned to a particular worker, or a computing device positioned at a location within a work environment, and at least one server is configured to receive the stream of usage data, store the at least one model, and detect the safety event signature in the stream of usage data based on processing the stream of usage data with the model.

41. The computing device of claim 40, wherein the one or more computer processors:

select at least one of the article of PPE, the worker device, the computing device, or the at least one server to detect the safety event signature in the stream of usage data.

42. The computing device of claim 41, wherein the selection occurs at the at least one of the article of PPE, the worker device, the computing device, or the at least one server to detect the safety event signature in the stream of usage data.

43. The computing device of claim 41, wherein the selection is based on a power consumption associated with detecting the safety event signature, a latency associated with detecting the safety event signature, a connectivity status of the article of PPE, the worker device, the computing device, or the at least one server, a data type of the usage data, a data volume of the usage data, and the content of the usage data.

44. The computing device of claim 29, wherein at least one sensor generates usage data that characterizes at least a worker associated with the article of PPE or a work environment and wherein detecting the safety event signature in the stream of usage comprises processing the usage data that characterizes the worker associated with the article of PPE or the work environment.

45. The computing device of claim 29, wherein to generate the output in response to detecting the safety event signature, the one or more computer processors send a notification to at least one of the article of PPE, a hub associated with a user and configured to communicate with the article of PPE and at least one remote computing device, or a computing device associated with person who is not the user.

46. The computing device of claim 29, wherein to generate the output in response to detection of the safety event signature, the one or more computer processors send a notification that alters an operation of the article of PPE.

47. The computing device of claim 29, wherein to generate the output in response to detection of the safety event signature the one or more computer processors output for display a user interface that indicates the safety event in association with at least one of a user, work environment, or the article of PPE.

48. The computing device of claim 29, wherein the article of PPE comprises at least one of an air respirator system, a fall protection device, a hearing protector, a head protector, a garment, a face protector, an eye protector, a welding mask, or an exosuit.

49. The computing device of claim 29, wherein the article of PPE is included in a set of articles of PPE associated with a user, wherein each article of PPE in the set of articles of PPE includes a motion sensor, wherein the one or more computer processors:

receive a respective stream of usage data from each respective motion sensor of each respective article of PPE of the set of articles of PPE; and

to detect the safety event signature, the one or more computer processors detect the safety event signature corresponding to a relative motion that is based at least in part on the respective stream of usage data from each respective motion sensor.

50. The computing device of claim 29, wherein the stream of usage data comprises events, wherein each respective event is generated at a same defined interval, wherein each respective event includes a respective set of values that correspond to a same set of defined metrics, and wherein respective sets of values in different respective events are different.

51. The computing device of claim 50, wherein the set of defined metrics comprises one or more of a timestamp, characteristics of the article of PPE, characteristics of a worker associated with the article of PPE, or characteristics a work environment.

52. The computing device of claim 29, wherein at least one safety rule is mapped to at least one safety event, wherein the at least one safety event is mapped to the safety event signature, and wherein the safety event signature corresponds to at least the portion of the stream of usage data.

53. The computing device of claim 29, wherein to detect the safety event signature in the stream of usage data based on processing the stream of usage data with the model, the one or more computer processors determine a set of one or more likelihoods associated with one or more safety event signatures, wherein the safety event signature is associated with a likelihood in the set of one or more likelihoods associated with one or more safety event signatures.

54. The computing device of claim 29, wherein to generate an output in response to detecting the safety event signature, the one or more processors generate a user interface that is based at least in part on a safety event that corresponds to the safety event signature.

55. The computing device of claim 54, wherein the user interface includes at least one input control that requires a responsive user input within a threshold time period, wherein the one or more computer processors:

in response to the threshold time period expiring without the responsive user input, perform at least one operation based at least in part on the threshold time period expiring without the responsive user input.

56. The computing device of claim 29, wherein the safety event signature corresponds to a safety event that indicates ergonomic stress that satisfies a threshold.