Several embodiments include a portable security device. The portable security device can include one or more sensors. The portable security device can compute a home rhythm pattern utilizing a machine learning engine based on a historical record of real-time sensor feeds. The portable security devic
Several embodiments include a portable security device. The portable security device can include one or more sensors. The portable security device can compute a home rhythm pattern utilizing a machine learning engine based on a historical record of real-time sensor feeds. The portable security device can camouflage itself as a digital clock, a digital calendar, or a home security dashboard. The portable security device can define an action trigger that binds a state of the environment around the portable security device to at least a device component action. The portable security device can identify a real-time state of the portable security device amongst a finite set of potential states based on features observed from the sensor feeds. The portable security device can execute the device component action at the portable security device in response to determining that the real-time state matches the action trigger.
대표청구항▼
1. A method of operating a portable security device comprising: monitoring one or more sensor feeds in real-time from one or more sensors in the portable security device, the one or more sensor feeds providing information indicative of activity occurring within an environment of the portable securit
1. A method of operating a portable security device comprising: monitoring one or more sensor feeds in real-time from one or more sensors in the portable security device, the one or more sensor feeds providing information indicative of activity occurring within an environment of the portable security device;computing a home rhythm pattern utilizing a machine learning engine based on a historical record of the sensor feeds, the computing of the home rhythm pattern including determining that the historical record of the sensor feeds satisfies a statistical threshold related to activity occurring within the environment;defining an action trigger based on the home rhythm pattern, wherein the action trigger specifies a state of an environment of the portable security device amongst a finite set of potential states, wherein the action trigger binds at least a device component action to the specified state;identifying a real-time state of the environment of the portable security device based on features observed from the sensor feeds;executing the device component action at the portable security device in response to determining that the real-time state matches the action trigger;determining a response of a user within the environment of the portable security device to the execution of the device component action at the portable security device;adjusting the machine learning engine to provide an adjusted machine learning engine based on the response of the user within the environment of the portable security device to the execution of the device component action at the portable security device: andcomputing a second home rhythm pattern utilizing the adjusted machine learning engine and based on the historical record of the sensor feeds, the second home rhythm pattern being different than the home rhythm pattern. 2. The method of claim 1, further comprising verifying the home rhythm or the action trigger via a control interface displayed on a touch screen of the portable security device. 3. The method of claim 1, wherein the home rhythm pattern includes an environmental state change pattern or a user behavior pattern. 4. The method of claim 1, wherein executing the device component action includes configuring an output component device of the portable security device, the output component device being a display, a speaker, a light, or any combination thereof. 5. The method of claim 1, wherein executing the device component action includes recording an event utilizing at least an input component device of the portable security device. 6. The method of claim 1, wherein the sensors include a temperature sensor, an air quality sensor, a humidity sensor, a microphone, a motion detector, a camera, or any combination thereof. 7. The method of claim 1, wherein executing the device component action includes configuring an external device utilizing a network interface of the portable security device. 8. The method of claim 7, wherein the external device is an Internet of Things (loT) device, a computing device, a cloud computing system, or any combination thereof. 9. The method of claim 1, wherein executing the device component action includes sending a message via a network interface of the portable security device, wherein the message includes natural language audio or textual content. 10. The method of claim 1, further comprising providing access to a control interface to monitor or edit the action trigger or the home rhythm via an application programming interface (API) to a user device, an API to a third-party system, an API to a locally-implemented web server, an API to a cloud server, or any combination thereof. 11. The method of claim 1, wherein identifying the real-time state is by determining that a measurement reading from the sensor feeds is within a preset range associated with the real-time state, and wherein at least one of the potential states has a corresponding preset range. 12. The method of claim 1, wherein identifying the real-time state is based on one or more statistical characteristics of the sensor feeds. 13. The method of claim 12, wherein the statistical characteristics include a moving average, a moving variance, a moving standard deviation, a current deviation from the moving average, or any combination thereof. 14. The method of claim 1, wherein identifying the real-time state is based on machine recognition of a known pattern in the sensor feeds. 15. The method of claim 1, wherein identifying the real-time state is based on machine recognition of an unknown pattern in the sensor feeds utilizing an unsupervised machine learning algorithm. 16. The method of claim 1, wherein identifying the real-time state is based on determining whether the sensor feeds indicate a deviation beyond a threshold amount or percentage from baseline values according to the home rhythm. 17. The method of claim 1, further comprising: monitoring one or more external data feeds in real-time from devices in a local area network to which the portable security device is connected; andwherein computing the home rhythm pattern is based on a combination of the historical record of the sensor feeds and the external data feeds. 18. The method of claim 17, wherein identifying the real-time state is based on features observed from the sensor feeds, the external data feeds, or a combination thereof. 19. A portable security device comprising: a touchscreen display covering or substantially covering at least one side of the portable security device, and wherein the touchscreen display is configured to present an interactive control interface;one or more sensors configured to provide one or more sensor feeds of sensor readings, wherein the sensors include at least a temperature sensor, an air quality sensor, a motion detection sensor, or a humidity sensor;a network connection interface configured to connect with a local area network (LAN), a peer-to-peer device, a wide area network (WAN), or any combination thereof;a processor configured to: compute a home rhythm pattern utilizing a machine learning engine based on a historical record of the sensor feeds, wherein the home rhythm pattern includes an environmental state change pattern or a user behavior pattern, the computing of the home rhythm pattern including determining that a historical record of the environmental state change pattern or the user behavior pattern satisfies a statistical threshold related to activity occurring within an environment of the portable security device;define an action trigger that specifies a state of an environment of the portable security device, wherein the state is selected from amongst a finite set of potential states;identify a real-time state of the environment of the portable security device based on numeric features observed in the sensor feeds;execute a device component action at the portable security device in response to determining that the real-time state matches the action trigger;determine a response of a user within the environment of the portable security device to the execution of the device component action at the portable security device;adjusting the machine learning engine to provide an adjusted machine learning engine based on the response of the user within the environment of the portable security device to the execution of the device component action at the portable security device; andcomputing a second home rhythm pattern utilizing the adjusted machine learning engine and based on the historical record of the sensor feeds, the second home rhythm pattern being different than the home rhythm pattern. 20. The portable security device of claim 19, further comprising a shell protecting the network connection interface, the processor, and the sensors. 21. The portable security device of claim 20, and wherein the shell exposes at least a hole over at least one of the sensor for the one sensor to observe the environment of the portable security device. 22. The portable security device of claim 20, and wherein the shell is attached to the touchscreen display and has a hole on the side across from the touchscreen display to enable the portable security device to be hung on a wall. 23. A non-transitory computer readable medium storing instructions there on which, when executed by a processor, cause the processor to: monitor one or more sensor feeds in real-time from one or more sensors in a portable security device, the one or more sensor feeds providing information indicative of activity occurring within an environment of the portable security device;compute a home rhythm pattern utilizing a machine learning engine based on a historical record of the sensor feeds, the computing of the home rhythm pattern including determining that the historical record of the sensor feeds satisfies a statistical threshold related to activity occurring within the environment;define an action trigger based on the home rhythm pattern, wherein the action trigger specifies a state of an environment of the portable security device amongst a finite set of potential states, wherein the action trigger binds at least a device component action to the specified state;identify a real-time state of the environment of the portable security device based on features observed from the sensor feeds;execute the device component action at the portable security device in response to determining that the real-time state matches the action triggerdetermine a response of a user within the environment of the portable security device to the execution of the device component action at the portable security device;adjust the machine learning engine to provide an adjusted machine learning engine based on the response of the user within the environment of the portable security device to the execution of the device component action at the portable security device; andcompute a second home rhythm pattern utilizing the adjusted machine learning engine and based on the historical record of the sensor feeds, the second home rhythm pattern being different than the home rhythm pattern. 24. The method of claim 1, wherein executing the device component action at the portable security device in response to determining that the real-time state matches the action trigger includes adjusting functionality of a device within the environment of the portable security device, the functionality of the device adjusted to correspond to functionality of the device as indicated in the historical record of the sensor feeds. 25. The method of claim 1, further comprising: defining the action trigger based on the second home rhythm pattern, wherein the action trigger based on the second home rhythm pattern specifies a second state of an environment of the portable security device that is different than the state of the environment of the portable security device specified by the home rhythm pattern. 26. The method of claim 1, further comprising: determining environmental context conditions of the environment of the portable security device, wherein the action trigger is also based on the environmental context conditions of the environment. 27. The method of claim 1, wherein the response of the user to the execution of the device component action includes the user providing an indication that the home rhythm pattern should be adjusted.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (18)
Bousfield, Randall Dennis; Padilla, Toby Mark, Access control reader.
Yoshiike, Nobuyuki; Hattori, Akiyoshi; Morinaka, Katsuya; Inoue, Shigeyuki; Tanaka, Shinji, Behavior determining apparatus, care system, care residence and behavior information specifying apparatus and system.
Fisher, Scott R.; Caskey, Matthew K.; Crutcher, John P.; Glassburn, Robin S.; Morris, Wendell H., Electronic lock box with mechanism immobilizer features.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.