Context awareness by intelligent devices sensing transient and continuous events

Abstract

Embodiments of the present disclosure relate to context awareness by intelligent devices that sense both transient events and continuous events. A distributed computing system is used to autonomously understand artificial intelligence in the environmental context of an intelligent device. Raw context data is detected by sensors associated with the intelligent device. The raw context data is pre-processed by the intelligent device and then provided to a cloud-based server for further processing. At the cloud-based server, various feature data sets are derived from the pre-processed context data. The various feature data sets are compared to corresponding classification parameters to determine a classification of a continuous event and / or a classification of a transient event (if any) that occurred in the context. The determined classification of the continuous event and the transient event will be used to autonomously configure the intelligent device or another related intelligent device to adapt to the context.

Claims

1. A system for configuring intelligent devices, comprising: One or more sensors, the one or more sensors being configured to detect context data including at least one of motion, sound and spatial information of a context; as well as A storage medium containing executable instructions, which, when executed by a processing unit, configure the processing unit to perform actions, the actions including: Receive the context data from one or more sensors. Generate a first feature data set from the received context data. Based on the first feature data set, it is determined that a transient event occurred in the context. A second feature data set is generated from the received contextual data. The classification of the transient event is determined based on the second feature data set. Generate a third feature data set from the received contextual data. The classification of continuous events in the context is determined based on the third feature data set. Based on the determined classification of the transient events and the determined classification of the continuous events, the intelligent device having processor circuitry and associated with one or more sensors in the context is configured; and Generating the second feature data set includes processing the received context data using a first series of time windows; and Generating the third feature data set includes processing the received context data using different second series of time windows.

2. The system of claim 1, wherein the classification of the transient event is in at least one of a motion vector, a sound vector, and a spatial environment vector.

3. The system according to claim 1, wherein generating the first feature data set comprises: Use a series of time windows to process the received contextual data.

4. The system according to claim 1, wherein the time window in the first series of time windows is shorter than the time window in the second series of time windows.

5. The system of claim 1, wherein determining the classification of the transient event based on the second feature data set comprises: The second feature dataset is analyzed using multiple classification parameters.

6. The system of claim 5, wherein the analysis comprises posterior graphical probability analysis of the second feature data set relative to a classification parameter library.

7. The system of claim 6, wherein the classification parameter library is customized for at least one of user, geographic location, or time.

8. The system of claim 1, wherein the generation of the feature data set is performed for contextual information detected by each of the one or more sensors.

9. The system according to claim 1, further comprising: The classification of continuous events and the classification of transient events are combined to obtain meta-level context-aware analysis results.

10. The system of claim 9, wherein the classification of the continuous events and the classification of the transient events comprise: Remove at least one of the classifications of the continuous events and the transient events.

11. The system of claim 9, wherein the classification of the continuous events and the classification of the transient events comprise: The classification of transient events in one of the motion vector, sound vector, and spatial environment vector is combined with transient events in another of the motion vector, sound vector, and spatial environment vector.

12. The system according to claim 1, further comprising: The received contextual data is filtered based on the characteristics of human activities within the context.

13. The system of claim 1, wherein the classification for determining the transient event takes into account consecutive events preceding the transient event.

14. The system according to claim 1, wherein: The classification of the transient event is determined in a classification element included in at least one of the first motion vector, the first sound vector, and the first spatial environment vector; The classification of the continuous events is determined from classification elements included in at least one of the second motion vector, the second sound vector, and the second spatial environment vector; and The first motion vector and the second motion vector repel each other at the same point in time in the situation.

15. The system of claim 1, wherein the intelligent device configured to be associated with the one or more sensors in the context comprises: The intelligent device is configured to automatically perform actions based on the determined classification of the transient events and the determined classification of the continuous events.

16. A system for configuring intelligent devices, comprising: One or more sensors, the one or more sensors being configured to detect context data including at least one of motion, sound and spatial information of a context; as well as A non-transitory storage medium containing executable instructions, which, when executed by a processing unit, configure the processing unit to perform actions, the actions including: Receive the context data from one or more sensors. Generate a first feature data set from the received context data. Based on the first feature data set, it is determined that a transient event occurred in the context. A second feature data set is generated from the received contextual data. The classification of the transient event is determined based on the second feature data set. Generate a third feature data set from the received contextual data. The classification of continuous events in the context is determined based on the third feature data set. Combining the classifications of the continuous events and the classifications of the transient events yields meta-level context-aware analysis results, and Based on the determined classification of the transient events and the determined classification of the continuous events, including based on the meta-level context-aware analysis results, the intelligent device having processor circuitry and associated with one or more sensors in the context is configured.

17. The system of claim 16, wherein the action includes determining a sequence of context updates based on a determined classification of the transient event and a determined classification of the continuous event, and The configuration of the smart device is based on the sequence of context updates.

18. The system of claim 16, wherein the classification of the transient event is in at least one of a motion vector, a sound vector, and a spatial environment vector.

19. A system for configuring intelligent devices, comprising: One or more sensors, the one or more sensors being configured to detect context data including at least one of motion, sound and spatial information of a context; as well as A storage medium containing executable instructions, which, when executed by a processing unit, configure the processing unit to perform actions, the actions including: Receive the context data from one or more sensors. Generate a first feature data set from the received context data. Based on the first feature data set, it is determined that a transient event occurred in the context. A second feature data set is generated from the received contextual data. The classification of the transient event is determined based on the second feature data set. Generate a third feature data set from the received contextual data. The classification of continuous events in the context is determined based on the third feature data set, and Based on the determined classification of the transient events and the determined classification of the continuous events, the intelligent device, having processor circuitry and associated with one or more sensors in the context, is configured. in: The classification of the transient event is determined in a classification element included in at least one of the first motion vector, the first sound vector, and the first spatial environment vector; The classification of the continuous events is determined from classification elements included in at least one of the second motion vector, the second sound vector, and the second spatial environment vector; and The first motion vector and the second motion vector repel each other at the same point in time in the situation.

20. The system of claim 19, wherein determining the classification of the transient event based on the second feature data set comprises: The second feature dataset is analyzed using multiple classification parameters.

Images