Artificial Intelligence for Internet of Things Insights from Patents
0 likes507 views
The document discusses the integration of artificial intelligence (AI) in Internet of Things (IoT) applications, highlighting how patents can reveal innovations in this area. It details a smart home intelligence system that personalizes appliance settings based on user behaviors and preferences, as well as examples of AI use in healthcare and connected cars. The document illustrates various patent examples that showcase how AI can analyze contextual data to enhance user experiences across different domains.
Artificial Intelligence for Internet of Things Insights from Patents
- 1. ©2015 TechIPm, LLC All Rights Reserved http://www.techipm.com/ 1 Artificial Intelligence for Internet of Things Insights from Patents Artificial Intelligence (AI) is essential to provide value added Internet of Things (IoT) services by finding the patterns, correlations and anomalies in user behaviors for autonomous context-aware actions of the IoT system surrounding the user. Patents can provide insights regarding the state of the art and technical details of the AI innovation for the IoT applications. For example, US20150039105 (Home intelligence system and method; Wen- Sung Lee) illustrates the smart home intelligence system to fulfill the special needs of each family member exploiting AI. The smart home intelligence system includes an identification sensor, a real-time locating system and an AI device. The identification sensor is configured to detect presence of a home member and identify ID information of the home member. The real-time locating system is provided to automatically identify and track location of the home member in real time within the house. The AI device includes a data logger, a controller and a learning module. The data logger is provided to record the ID information of the home member from the identification sensor and further records location information of the home member from the real-time locating system. The controller is configured to detect control parameters of appliances switched or adjusted by the home member and transmits the detected control parameters to the data logger. The learning module is configured to correlate at least the ID information and the location information with the detected control parameters to customize for the home member a
- 2. ©2015 TechIPm, LLC All Rights Reserved http://www.techipm.com/ 2 set of preferred control parameters for operating the appliances. Accordingly, upon the home member is detected with the ID information and the location information, the controller retrieves from the learning module the set of preferred control parameters based on the ID information and the location information and directs the appliances to operate with the set of preferred control parameters for the detected home member. For example, a family member (A) enters in the first floor room, and therefore the identification sensor detects the presence of the family member (A) and identifies by the profile of the family member that the family member (A) is a father with ID information (father), the real-time locating system automatically identifies and tracks location of the father in real time within the house. The identification sensor, the real-time locating system and the time clock then transmit the ID information, the location information and the time information to the data logger. The data logger receives and records the father's ID information from the identification sensor, the location information from the real-time locating system and the time information from the time clock. Father then utilizes his smartphone to adjust the temperature of the HVAC system 61 to be 22° C., the lighting system to provide an illuminance of 50 lux, and the stereo system to be off. At this time, the controller detects the control parameters of appliances adjusted by the father and transmit the detected control parameters to the data logger. The learning module then correlates the ID information (father), the location information (1st floor room), and the time information (7:00 am) with the detected control parameters to customize for the home member a set of preferred control parameters for operating the appliances. In this way, next time the father comes in the first floor room again, the father will be detected by the home intelligence system and the system will automatically directs the HVAC system to work on
- 3. ©2015 TechIPm, LLC All Rights Reserved http://www.techipm.com/ 3 22° C., the lighting system 62 to provide an illuminance of 50 lux, and the stereo system 63 to be off. In this case, the learning module selects, among different previously used control parameters, a set of most recently used control parameters as the set of preferred control parameters for the home member. The learning module may select a set of his most frequently used control parameters as the set of preferred control parameters for the home member. AI can also be exploited in the smart healthcare applications. For example, wearable devices track your health condition, e.g. heart rate, and send a notification if a potential problem is detected based on AI analysis of the real- time streams of health data. For example, US20140073486 (SYSTEMS, DEVICES AND METHODS FOR CONTINUOUS HEART RATE MONITORING AND INTERPRETATION: BOBO ANALYTICS) illustrates a heart rate monitoring system by providing the best type of sensor to use at a given time determined by AI based on the level of motion (e.g., via an accelerometer) and whether the user is asleep (e.g., based on movement input, skin temperature and heart rate). Common symptoms of being asleep are periods of no movement or small bursts of movement (such as shifting in bed), lower skin temperature (although it is not a dramatic drop from normal), and heart rate that is below the typical resting heart rate when the user is awake. These variables depend on the physiology of a person and thus a machine learning algorithm is trained with user-specific input to determine when he/she is awake/asleep and determine from that the exact parameters that cause the algorithm to deem someone asleep.
- 4. ©2015 TechIPm, LLC All Rights Reserved http://www.techipm.com/ 4 US20140108307 (METHODS AND SYSTEMS FOR PROVIDING PERSONALIZED AND CONTEXT- AWARE SUGGESTIONS; Wipro) illustrates AI exploitation in the connected car applications: Base on the profile information and/or contextual information, AI system provides suggestions to the driver. For instance, the AI system obtains contextual information indicating that a contact of the driver is only a few minutes in driving time from the driver’s route. Further, based on social communications and social media information stored in the driver profile, the AI system may also determine that the contact is a friend of the user. Based on that determination, the AI system informs the driver, "Your friend Peter is at a few minutes away from your route; would you like meet?"