Button-Less Vaping Gesture Detection for Low-Latency Preheating
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic vaping devices lack an efficient method to detect hand-to-mouth gestures, leading to delayed vapor production and an unsatisfactory user experience due to the need for manual activation.
Innovation Solution
A method utilizing device sensors, such as gyroscopes and accelerometers, to detect movements and generate quaternions, which are then transformed into Cartesian coordinates and filtered to extract movement features. These features are applied to a classifier trained with linear discriminant analysis to determine hand-to-mouth gestures, enabling automatic transition of the heater's operation mode from a low power state to a higher power state for preheating the pre-vapor formulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual activation is used to control vapor production, then device complexity is reduced, but user experience deteriorates due to delayed vapor production
Solution Approach 1:
The system performs preliminary action by detecting the hand-to-mouth gesture and initiating heater preheating before the user actually needs vapor. The classifier processes movement data in real-time to anticipate user intent, allowing the heating element to prepare in advance, thus reducing vapor latency when the user takes a puff.
2Loss of time
If gesture detection system is implemented, then vapor production timing is improved, but device complexity increases due to additional sensors and processing
Solution Approach 1:
The device integrates multiple functions into existing components: the accelerometer and gyroscope serve both for gesture detection and for detecting device orientation/position. The processor handles both vapor control logic and gesture classification tasks. This multi-functionality approach minimizes additional hardware requirements while achieving real-time gesture recognition.
Solution Approach 2:
The patent introduces a classifier as an intermediary layer between raw sensor data and control decisions. The classifier processes quaternion data from accelerometer and gyroscope readings, transforming complex sensor inputs into interpretable gesture classifications. This intermediary simplifies the control logic by providing a standardized interface between sensing and actuation systems.
3Speed
If real-time gesture classification is performed, then responsiveness is improved, but energy consumption increases due to continuous processing
Solution Approach 1:
The system employs periodic action by updating the classifier with quaternion data at specific intervals rather than continuously. The processor calculates quaternions from accelerometer and gyroscope readings periodically, feeds them to the classifier when needed, and maintains heater control states between updates. This periodic processing approach balances real-time responsiveness with energy conservation.
Data Source
AI summary
A method of detecting a hand-to-mouth (HMG) gesture with an e-vaping device includes detecting movements of the e-vaping device; generating quaternions based on the detected movements; generating movement features based on the generated quaternions; applying the generated movement features to a classifier; and determining whether the detected movements correspond to an HMG based on an output of the classifier.