Other

Newvectors LLC

Techteam Government Solutions, Inc.

Sven Brueckner

Henry Van Dyke Parunak

H Van Dyke Parunak

John A Sauter

Peter Michael Weinstein

Paul Chiusano

Robert J Bisson

Robert S Matthews

Theodore C Belding

Steven M Brophy

A clustering method yields a searchable hierarchy to speed retrieval, and can function dynamically with a changing document population. Nodes of the hierarchy climb up and down the emerging hierarchy based on locally sensed information. Like previous ant clustering algorithms, the inventive process is dynamic, decentralized, and anytime. Unlike them, it yields a hierarchical structure. For simplicity, and reflecting our initial application in the domain of textual information, the items being clustered are documents, but the principles may be applied to any collection of data items.

Hierarchical ant clustering and foraging

A method of predicting the behavior of software agents in a simulated environment involves modeling a plurality of software agents representing entities to be analyzed, which may be human beings. Using a set of parameters that governs the behavior of the agents, the internal state of at least one of the agents is estimated by its behavior in the simulation, including its movement within the environment. This facilitates a prediction of the likely future behavior of the agent based solely upon its internal state; that is, without recourse to any intentional agent communications. In the preferred embodiment the simulated environment is based upon a digital pheromone infrastructure. The simulation integrates knowledge of threat regions, a cognitive analysis of the agent's beliefs, desires, and intentions, a model of the agent's emotional disposition and state, and the dynamics of interactions with the environment. By evolving agents in this rich environment, we can fit their internal state to their observed behavior. In realistic wargame scenarios, the system successfully detects deliberately played emotions and makes reasonable predictions about the entities' future behavior.

Characterizing and predicting agents via multi-agent evolution

A continuous, emergent, anytime process clusters input documents according to a similarity function within a node-based, distributed computing environment, for example, within a client/server environment. An agent (DAg) assigned to each document determines whether its document should remain at a node or be moved to another node to increase similarity clustering. An agent (SAg) assigned to each node may be operative to manage storage requirements within its node, and/or manage communications between the nodes of the environment as the DAgs operate. Typically a move request is issued to another node if it is determined that clustering would increase by moving a document to that node. In such an instance, the SAg assigned to that other node would probabilistically consider the move request in view of other such requests in sequence to avoid overloading. To enhance performance, documents may be preprocessed and given values representative of similarity. The similarity between two arbitrary documents may then assessed as the average difference between the values assigned to the same attribute in the two documents.

Agent-based clustering of abstract similar documents

A swarming agent architecture provides a distributed, decentralized, agent-based computing environment applicable to ground-based surveillance. The approach, called Sensor Network Integration through Pheromone Fusion, or 'SNIPF,' provides an end-to-end demonstration that integrates self-contained sensor/communication nodes with novel swarming algorithms to detect foot and vehicular movement through a monitored area with minimal configuration and maintenance. A plurality of computational nodes distributed within the environment and, depending upon the way in which they are deployed, the various nodes are operative to sense the local environment, receive a message from a neighboring node, and transmit a message to a neighboring node. Given these capabilities, the nodes can collectively determine the presence and/or movement of a target and communicate this information to a user. Though not required, the system may include nodes that are capable of collectively determining the speed and heading of a target, and the gathered intelligence may be communicated to users within, and external to, the environment. A particularly useful configuration may include one or more 'free' nodes having relatively limited communications and computational power, and one or more anchor nodes equipped with GPS and/or long-distance communications capabilities.

Decentralized detection, localization, and tracking utilizing distributed sensors

A system and process uses pheromone propagation to provide direction and/or distance information to a remote destination through the use of localized decision making processes. Locally available information is exchanged and translated with global information through the use of a pheromone vocabulary and multi-layer grid systems. Self-organizing networks of sensors provide remote monitoring of a space.

Spatial coordination system

An Adaptive, Any-Time Case Retrieval process combines existing knowledge (emergent clustering, case retrieval with CRN) in a novel and advantageous way. Although ideally suited to the Digital Body Development System (DBDS), the method is applicable to any problem solving environment involving iterative simulations. Given a description of a problem to be solved, a candidate solution is applied to generate a problem solving base case. The description of the problem is modified and the modification is recorded. Candidate solutions are applied to the problem as modified, thereby generating a problem solving event case. These steps are repeated so as to rank the relevance of the cases generated to arrive at an optimal solution to the problem. In the preferred embodiment a case retrieval network (CRN) structure is used in the ranking process.

Problem solving method utilizing emergent clustering and case retrieval

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.

Growing community of inventors

Dexter, MI, United States of America

Sven Brueckner