Mobile Context Vectors for Routine Prediction Without GPS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobile devices rely on location data to provide predictive user assistance, which is limited in situations where location information is unavailable, such as indoors, and requires manual labeling or surveys, making it inefficient for providing assistance in routine tasks across multiple locations.
Innovation Solution
A mobile device uses data-driven context determination by recording sensor readings from various sources, such as wireless signals, temperature, and sound, to create context vectors that can predict user actions and provide assistance without requiring geographic location information or manual labeling, allowing it to recognize events and transitions between environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If mobile devices use location data to provide predictive user assistance, then user assistance accuracy is improved, but the system fails to provide assistance in environments where location data is unavailable (e.g., indoors)
Solution Approach 1:
The patent segments the location determination problem into two parts: GPS-based location for outdoor environments and sensor-based context vectors for indoor environments. This segmentation allows the system to use appropriate methods for each environment, maintaining reliability outdoors while extending adaptability indoors where GPS is unavailable.
Solution Approach 2:
The patent introduces context vectors as an intermediary between physical environment and predictive assistance. These context vectors, derived from sensor readings (accelerometer, gyroscope, magnetometer, barometer, microphone, camera, Wi-Fi, Bluetooth), serve as mediators that enable the system to recognize environments and provide assistance without direct GPS location data.
2Ease of operation
If mobile devices require manual labeling or surveys to provide predictive assistance, then location-based assistance can be provided, but the process becomes inefficient and complex
Solution Approach 1:
The patent implements self-service by enabling the mobile device to automatically generate context vectors from its own sensor readings without requiring user input. The device autonomously determines environments, identifies routines, and provides predictive assistance, eliminating the need for manual labeling or surveys while reducing operational complexity.
Solution Approach 2:
The patent changes the parameters used for environment identification from manual location labels to automated sensor-based context vectors. By transforming physical sensor readings into contextual representations, the system automatically adapts to environments without requiring users to configure location data or perform surveys.
3Productivity
If mobile devices use conventional location-based prediction methods, then assistance can be provided for outdoor locations, but the system cannot recognize events or transitions between indoor environments
Solution Approach 1:
The patent makes the context vector system universal by designing it to handle both outdoor and indoor environments, as well as recognize both locations and events. The same sensor-based approach that identifies outdoor locations also identifies indoor environments, room transitions, and user actions, eliminating information loss across different environment types.
Solution Approach 2:
The patent adds a new dimension to location-based prediction by incorporating temporal and contextual dimensions through context vectors. Instead of relying solely on spatial GPS coordinates, the system uses multi-dimensional sensor data (motion, orientation, environmental sensors) to capture indoor environments and event transitions, recovering information that conventional location-based systems lose.
Data Source
AI summary
A mobile device can provide predictive user assistance based on various sensor readings, independently of or in addition to a location of the mobile device. The mobile device can determine a context of an event. The mobile device can store the context and a label of the event on a storage device. The label can be provided automatically by the mobile device or by the external system without user input. At a later time, the mobile device can match new sensor readings with the stored context. If a match is found, the mobile device can predict that the user is about to perform the action or recognize that the user has performed the action again. The mobile device can perform various operations, including, for example, providing user assistance, based on the prediction or recognition.