Swarming agents for distributed pattern detection and classification

• US 20030142851A1
• Filed: 01/27/2003
• Published: 07/31/2003
• Est. Priority Date: 07/06/2001
• Status: Active Grant

• Current Assignee: Techteam Government Solutions Incorporated (Jacobs Solutions, Inc.)
• Original Assignee: Techteam Government Solutions Incorporated (Jacobs Solutions, Inc.)
• Inventors: Parunak, H.Van Dyke, Brueckner, Sven
• Granted Patent: US 7,305,371 B2
• Time in Patent Office: Days
• US Class Current: 382/107
• CPC Class Codes: B25J 9/1617 Cellular, reconfigurable manipulator, e.g. cebot
• CPC Class Codes: G01S 5/0289 of multiple transceivers, e.g. in ad hoc networks
• CPC Class Codes: G01S 7/003 Transmission of data between radar, sonar or lidar systems and remote stations
• CPC Class Codes: G06F 17/10 Complex mathematical operations function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544
• CPC Class Codes: G06F 18/21 Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16
• CPC Class Codes: H04L 41/048 mobile agents
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: Y04S 40/18 Network protocols supporting networked applications, e.g. including control of end-device applications over a network

Abstract

Swarming agents in networks of preferably physically distributed processing nodes are used for data acquisition, data fusion, and control applications. An architecture for active surveillance systems is presented in which simple mobile agents collectively process real-time data from heterogeneous sources at or near the origin of the data. System requirements are specifically matched to the needs of a surveillance system for the early detection of large-scale bioterrorist attacks on a civilian population, but the same architecture is applicable to a wide range of other domains. The pattern detection and classification processes executed by the proposed system emerge from the coordinated activities of agents of two populations in a shared computational environment. Detector agents draw each other'"'"'s attention to significant spatio-temporal patterns in the observed data stream. Classifier agents rank the detected patterns according to their respective criterion. The resulting system-level behavior is adaptive, robust, scalable, and applicable to a wide variety of other situations, including surveillance, financial transactions, network diagnosis, power grid monitoring, and others.

Claims

1. A swarming agent architecture for the distributed detection and classification of spatio-temporal patterns, comprising:

a population of mobile agents in a network of processing nodes; and

a service agent at each node, enabling the mobile agents to perform the following functions;

a) share information on an indirect basis,

b) coordinate activities across the entire population, and

c) detect and classify one or more spatio-temporal patterns based upon the information sharing, coordinated activities, or both.

2. The architecture of claim 1, wherein the spatio-temporal patterns occur in a heterogeneous real-time data stream.

3. The architecture of claim 1, wherein new classification schemes are introduced at different times through new agent populations.

4. The architecture of claim 1, wherein one or more of the spatio-temporal patterns are associated with a biological population.

5. The architecture of claim 1, wherein one or more of the spatio-temporal patterns are associated with an environment.

6. The architecture of claim 1, wherein one or more of the spatio-temporal patterns are associated with an event.

7. The architecture of claim 1, wherein one or more of the spatio-temporal patterns are associated with a network.

8. The architecture of claim 7, wherein the network is related to communications or energy.

9. The architecture of claim 1, wherein one or more of the spatio-temporal patterns are associated with surveillance.

10. The architecture of claim 1, wherein the definition of a pattern depends on the particular application of the detection mechanism.

11. The architecture of claim 1, including detector agents that search for spatio-temporal structures and classifier agents that classify the detected patterns according to a predetermined classification scheme.

12. The architecture of claim 11, wherein the pattern classification scheme expresses a particular spatial or temporal characteristic.

13. The architecture of claim 11, wherein the detector agents are attracted to locations which other detectors identified as part of a potential pattern, and repelled by large concentrations of other detector agents.

14. The architecture of claim 1, including places to represent the structure of the problem space in which the agents coordinate their activities.

15. The architecture of claim 14, wherein the places represent a physical space.

16. The architecture of claim 14, wherein the places represent a temporal space.

17. The architecture of claim 14, wherein the places represent an abstract space.

18. The architecture of claim 14, further including a place agent at the location of the processing to gather data from local sources and for storage to provide ready access.

19. The architecture of claim 14, wherein the place agent provides application agents with a unified interface to the data streams covered by the place.

20. The architecture of claim 14, wherein, the agents are operative to change focus by moving to a neighboring place, or by incrementing or decrementing an internal variable.

21. The architecture of claim 1, further including a data provider giving mobile agents access to data characteristics of the segment of the domain represented by the processing node.

22. A computer system programmed to implement the architecture of claim 1.

23. A swarming agent architecture for the distributed detection and classification of spatio-temporal patterns, comprising:

a population of mobile agents in a network of processing nodes, including;

detector agents that search for spatio-temporal patterns, and

classifier agents that classify the detected patterns according to a predetermined classification scheme.

24. The architecture of claim 23, wherein the spatio-temporal patterns occur in a heterogeneous real-time data stream.

25. The architecture of claim 23, wherein new classification schemes are introduced at different times through new agent populations.

26. The architecture of claim 23, wherein one or more of the spatio-temporal patterns are associated with a biological population.

27. The architecture of claim 23, wherein one or more of the spatio-temporal patterns are associated with an environment.

28. The architecture of claim 23, wherein one or more of the spatio-temporal patterns are associated with an event.

29. The architecture of claim 23, wherein one or more of the spatio-temporal patterns are associated with a network.

30. The architecture of claim 29, wherein the network is related to communications or energy.

31. The architecture of claim 23, wherein one or more of the spatio-temporal patterns are associated with surveillance.

32. The architecture of claim 23, wherein the definition of a pattern depends on the particular application of the detection mechanism.

33. The architecture of claim 31, wherein the pattern classification scheme expresses a particular spatial or temporal characteristic.

34. The architecture of claim 23, wherein the detector agents are attracted to locations which other detectors identified as part of a potential pattern, and repelled by large concentrations of other detector agents.

35. The architecture of claim 23, including places to represent the structure of the problem space in which the agents coordinate their activities.

36. The architecture of claim 35, wherein the places represent a physical space.

37. The architecture of claim 35, wherein the places represent a temporal space.

38. The architecture of claim 35, wherein the places represent an abstract space.

39. The architecture of claim 35, further including a place agent at the location of the processing to gather data from local sources and for storage to provide ready access.

40. The architecture of claim 35, wherein the place agent provides application agents with a unified interface to the data streams covered by the place.

41. The architecture of claim 35, wherein, the agents are operative to change focus by moving to a neighboring place, or by incrementing or decrementing an internal variable.

42. A computer system programmed to implement the architecture of claim 25.