Hrl Laboratories, LLC

Mike Howard

David W Payton

Craig A Lee

Mike Daily

Bruce Hoff

Rajan Bhattacharyya

Michael J Daily

Described is a system for flexible cognitive perception and selection. A pre-processing recognition module filters and tags input data from an environment resulting in a tagged percept. The tagged percept is stored and associated with a knowledge frame by a memory module based on shared descriptors, resulting in an activated knowledge frame. A utility rating is then supplied to each activated knowledge frame based on a set of reward values by an evaluation module. The activated knowledge frames are sorted, compared, and evaluated for a goodness of fit between the utility ratings of the activated knowledge frames and the input data by a hypothesis module. A best hypothesis for a current situation in the environment is determined based on a current highest rated activated knowledge frame.

Framework for flexible cognitive perception and action selection

An apparatus configured to obtain, process, and relay data to a user in a coherent and useful manner. An active fiducial is equipped with an interface for receiving and transmitting data. The fiducial may transmit its position using a satellite-based position sensing device such as a GPS. Active fiducials may also be equipped with battery power pack regenerated with solar cells. Similarly, the fiducials can be equipped with at least one video camera or other device having a focal plane array and a computer software system, configured to recognize shapes. The fiducials may also be equipped with inductive coils or other means for sensing metal containing compounds. The active fiducials may be equipped with a gas chromatograph. The active fiducials may use a variety of propulsion means including motor driven tracks, motor driven wheels, propellers, or other device or a combination of devices.

Distributed display composed of active fiducials

The present invention provides a distributed motion prediction network including a plurality of nodes. The nodes in the network comprise a mechanism for detecting the presence of an object in an area around a node. A node that detects an object is termed a 'detecting node.' After a node has detected an object, it communicates a signal to local nodes that are local with the detecting node to inform the local nodes of the presence of the object in the area around the detecting node. After the local nodes receive the signal, they further propagate the signal to other nodes (local to the local nodes) so that information regarding the presence of the object is progressively propagated to nodes away from the detecting node and so that as the object moves, the propagated signal is used to predict a direction of travel of the object.

Distributed motion prediction network

A method, an apparatus, and a computer program product for facilitating local signaling among a plurality of agents are presented. An agentincluding a digital processora transmitterand a directional receivercomprising a plurality of angular regionsis configured to enable the selection of a signal from a plurality of signals received on the basis of local data such as the strength of the received signal. The local data may be incorporated into a cumulative data portion of a signal that is passed from local agentto local agentacross a plurality of agentsin order to provide a gradient across the plurality of agentsThe gradient could be used, for example, to determine a shortest path across the agents. The signaling techniques described can also be used to determine the distance and direction to objects in the signaling path.

Method and apparatus for signaling among a plurality of agents

A method and apparatus for computing properties of a physical environment is provided, using a plurality of agents forming a distributed network embedded within the environment. The method comprises determining an initiating agent transmitting a signal including a cumulative cost value to neighboring agents and processing the signal at each neighboring agent to augment the cumulative cost value with local information If multiple signals are received, determining which has the best cumulative cost value for generating a new signal then treating the neighboring agent as an initiating agent and transmitting the new signal to neighboring agents and retaining the best augmented cost value in memory Methods further include determining paths using shortest path computations, using dual gradients for aligning agents on a path between two reference agents, and discovering and converging agents on choke points.

Method and apparatus for terrain reasoning with distributed embedded processing elements

A method for controlling the movement of agents using local communications is provided. Generally, each agent maintains an optimal distance from other local neighbor agents by, for each agent selecting a local agent measuring the distance and angle to the agent performing a distance maintenance calculation and repeating the distance maintenance calculation for each local agent. In the distance maintenance calculation an attraction/repulsion map is used in order to determine whether an agent is attracted to or repelled from other agents. A motion vector is used to determine agent responses to the attraction or repulsion. Over time, the agents settle into a neutral configuration where each is optimally distanced from the other agents. Reference agents and leader agents can be designated to direct the movement of other agents, and agents can be designated as blocking beacons to repel other agents from undesirable areas.

Growing community of inventors

Westlake Village, CA, United States of America

Mike Howard