Wearable Device Actuation Classification via Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices, particularly mobile devices, face issues with inadvertent actuation due to unintended motions or forces, leading to undesired consequences such as accidental button presses or unintended actions, which can disrupt ongoing tasks like phone calls.
Innovation Solution
The implementation of a system that classifies actuation events as valid or invalid by analyzing sensory data from accelerometers, gyroscopes, and other sensors, operational modes, and user settings to determine whether an actuation was intended or not, allowing the device to prevent unintended operations and consequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the device uses motion sensors and classification algorithms to distinguish valid from invalid actuations, then the reliability of actuation detection is improved, but the device complexity increases
Solution Approach 1:
The actuation detection system is segmented into multiple independent components: motion sensors (accelerometers, gyroscopes), pressure sensors, classification algorithms, and contextual analysis modules. Each component handles a specific aspect of detection, allowing the system to achieve high reliability through distributed sensing while managing complexity through modular architecture.
Solution Approach 2:
A classification algorithm acts as an intermediary between raw sensor data and actuation determination. This intermediary layer processes sensor inputs, applies contextual rules, and generates classified actuation events, thereby improving detection accuracy while abstracting the complexity from the final decision-making process.
2Measurement precision
If the device analyzes multiple sensor data sources and contextual information to classify actuations, then the precision of distinguishing intended vs. unintended actuations is improved, but the processing time and energy consumption increase
Solution Approach 1:
The system performs preliminary classification of actuation events using pre-defined rules and thresholds before full contextual analysis. Common actuation types are quickly identified and processed through simplified pathways, reducing overall processing time while maintaining precision for complex cases that require deeper analysis.
Solution Approach 2:
The system applies partial analysis to routine actuations and excessive (full) analysis only when needed. By using a tiered approach where simple actuations are processed with minimal data and only ambiguous or critical events trigger comprehensive multi-sensor analysis, the system achieves high precision without excessive processing delays for all events.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method of operating an electronic device, and an electronic device, are disclosed in which occurrences of unintended operations of the electronic device, such as can occur in response to inadvertent actuations of actuators of the electronic device, are reduced. In one example embodiment, the method includes detecting, by at least one sensor of a wearable computing device, motion of the wearable computing device, wherein the wearable computing device comprises a touch screen assembly including a piezoelectric sensor and a touchable lens surface, wherein the at least one sensor is different than the piezoelectric sensor; determining, by at least one processor of the wearable computing device, whether an actuation of the piezoelectric sensor is invalid based at least in part on whether a component of the motion exceeds a predetermined threshold, the component of the motion being normal to a plane of the touchable lens surface; and in response to determining that the actuation is invalid, performing, by the at least one processor, at least one operation.