Framework of hierarchical sensory grammars for inferring behaviors using distributed sensors
First Claim
Patent Images
1. A method for identifying a human behavior of an entity that takes place in space and time, comprising:
- a) receiving sensor data related to an action sequence of the entity from a pluralityof sensors that are positioned at respective nodes of a sensor network;
b) interpreting the sensor data according to a grammar hierarchy to produce a current set of semantics;
c) determining if a higher level grammar hierarchy exists;
d) interpreting the current set of semantics according to the higher level grammar hierarchy to produce a new set of semantics, wherein the new set of semantics comprises human behaviors of the current set of semantics that are interpreted as higher order human behaviors that take place at a more macroscopic level than the scale of the human behaviors of the current set of semantics;
e) identifying the new set of semantics as the current set of semantics;
f) repeating steps c, d and e, until it is determined that no higher level grammar hierarchy exists; and
g) outputting the current set of semantics as indicative of a particular human behavior of the entity,wherein each grammar level for each node generates outputs from the sensor data generated by each node, wherein the current set of semantics for each node is propagated across the sensor network; and
wherein each grammar hierarchy level computes a set of probabilities for each human behavior such that lower level grammar hierarchies infer simple human behaviors that allow for higher level grammar hierarchies to infer higher order human behaviors that take place at a more macroscopic level.
3 Assignments
0 Petitions
Accused Products
Abstract
Provided herein are methods, systems, and apparatuses that can utilize a grammar hierarchy to parse out observable activities into a set of distinguishable actions.
26 Citations
46 Claims
-
1. A method for identifying a human behavior of an entity that takes place in space and time, comprising:
-
a) receiving sensor data related to an action sequence of the entity from a pluralityof sensors that are positioned at respective nodes of a sensor network; b) interpreting the sensor data according to a grammar hierarchy to produce a current set of semantics; c) determining if a higher level grammar hierarchy exists; d) interpreting the current set of semantics according to the higher level grammar hierarchy to produce a new set of semantics, wherein the new set of semantics comprises human behaviors of the current set of semantics that are interpreted as higher order human behaviors that take place at a more macroscopic level than the scale of the human behaviors of the current set of semantics; e) identifying the new set of semantics as the current set of semantics; f) repeating steps c, d and e, until it is determined that no higher level grammar hierarchy exists; and g) outputting the current set of semantics as indicative of a particular human behavior of the entity, wherein each grammar level for each node generates outputs from the sensor data generated by each node, wherein the current set of semantics for each node is propagated across the sensor network; and
wherein each grammar hierarchy level computes a set of probabilities for each human behavior such that lower level grammar hierarchies infer simple human behaviors that allow for higher level grammar hierarchies to infer higher order human behaviors that take place at a more macroscopic level. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A method for determining a behavior of an entity that takes place in space and time, comprising:
-
receiving sensor measurements from a plurality of sensors that are positioned at respective nodes of a sensor network, wherein the sensor measurements are related to a spatial position or context of an entity; interpreting the sensor measurements as phonemes according to a phoneme definition; interpreting the phonemes according to a first grammar hierarchy to produce a first set of semantics, wherein the first grammar hierarchy comprises a plurality of human behaviors associated with a plurality of spatial positions; and interpreting the first set of semantics according to a second grammar hierarchy to produce a second set of semantics, wherein the second set of semantics is indicative of the human behavior of the entity that takes place at a more macroscopic level than the scale of the human behaviors of the first set of semantics, wherein the first and second grammar hierarchy for each node generates outputs from the sensor data generated by each node, wherein the second set of semantics for each node is propagated across the sensor network; and
wherein each grammar hierarchy computes a set ofprobabilities for each human behavior such that first level grammar hierarchies infer simple human behaviors that allow for the second grammar hierarchies to infer higher order human behaviors that take place at a more macroscopic level. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25)
-
-
26. A system for identifying a behavior of an entity that takes place in space and time, comprising:
-
a sensor; a memory; a processor, coupled to the sensor and the memory, wherein the processor is configured for performing the steps of; a) receiving sensor data related to an action sequence of the entity from the sensor and storing the sensor data on the memory; b) interpreting the sensor data according to a grammar hierarchy to produce a current set of semantics; c) determining if a higher level grammar hierarchy exists; d) interpreting the current set of semantics according to the higher level grammar hierarchy to produce a new set of semantics, wherein the new set of semantics comprises human behaviors of the current set of semantics that are interpreted as higher order human behaviors that take place at a more macroscopic level than the scale of the human behaviors of the current set of semantics; e) identifying the new set of semantics as the current set of semantics; f) repeating steps c, d and e, until it is determined that no higher level grammar hierarchy exists; and g) outputting the current set of semantics as indicative of a particular human behavior of the entity, wherein each grammar level for each node generates outputs from the sensor data generated by each node, wherein the current set of semantics for each node is propagated across the sensor network; and
wherein each grammar hierarchy level computes a set of probabilities for each human behavior such that lower level grammar hierarchies infer simple human behaviors that allow for higher level grammar hierarchies to infer higher order human behaviors that take place at a more macroscopic level. - View Dependent Claims (27, 28, 29, 30, 31, 32, 33, 34, 35, 36)
-
-
37. A computer readable medium with computer executable instructions embodied thereon for performing the steps comprising:
-
a) receiving sensor data related to an action sequence of the entity from a sensor; b) interpreting the sensor data according to a grammar hierarchy to produce a current set of semantics; c) determining if a higher level grammar hierarchy exists; d) interpreting the current set of semantics according to the higher level grammar hierarchy to produce a new set of semantics, wherein the new set of semantics comprises human behaviors of the current set of semantics that are interpreted as higher order human behaviors that take place at a more macroscopic level than the scale of the human behaviors of the current set of semantics; e) identifying the new set of semantics as the current set of semantics; f) repeating steps c, d and e, until it is determined that no higher level grammar hierarchy exists; and g) outputting the current set of semantics as indicative of a particular human behavior of the entity, wherein each grammar level for each node generates outputs from the sensor data generated by each node, wherein the current set of semantics for each node is propagated across the sensor network; and
wherein each grammar hierarchy level computes a set of probabilities for each human behavior such that lower level grammar hierarchies infer simple human behaviors that allow for higher level grammar hierarchies to infer higher order human behaviors that take place at a more macroscopic level. - View Dependent Claims (38, 39, 40, 41, 42, 43, 44, 45, 46)
-
Specification