Multi-Sensor Recognition Device for Automatic Mode Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current recognition devices for user actions, such as those mounted on the body to control devices like PCs or televisions, face challenges in reducing user operation burden by accurately distinguishing between joystick and touchpad actions, leading to increased steps in switching between modes and user fatigue.
Innovation Solution
A recognition device and method that utilize a combination of sensors on different body parts, like the wrist and finger, to acquire and process signals, calculating features based on these signals to accurately recognize joystick or touchpad actions, thereby automatically switching between action modes without user intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a recognition device uses sensors on different body parts to distinguish between joystick and touchpad actions, then the accuracy of action recognition is improved, but the device complexity increases
Solution Approach 1:
The recognition device divides the body into multiple segments (first body part and second body part) and places sensors on each segment. This segmentation allows the system to capture different aspects of the user's action - the first sensor captures motion at one location while the second sensor captures motion at another location, enabling differentiation between joystick and touchpad actions through comparative analysis of the segmented body parts' movements
Solution Approach 2:
The patent adds a spatial dimension to action recognition by utilizing the relative positional relationship between two different body parts. Instead of relying on a single sensor location, the system incorporates the spatial configuration and relative motion between multiple body segments, creating a multi-dimensional feature space that enables more accurate discrimination between action types
2Ease of operation
If the recognition device automatically switches between action modes, then the user operation burden is reduced, but the reliability of mode switching may decrease due to incorrect recognition
Solution Approach 1:
The recognition device implements a feedback mechanism where the processor continuously monitors the relative positional relationship between the first and second body parts, calculates features from sensor data, and adjusts mode switching decisions based on this feedback. The system uses the observed motion patterns and positional changes to dynamically determine whether to switch between joystick and touchpad modes, creating a closed-loop control system that adapts to the user's actual actions
3Measurement precision
If the system calculates features based on relative positional relationship changes, then the distinction between action types is improved, but the computational requirements increase
Solution Approach 1:
The patent extracts only the essential features needed for action recognition - specifically, the relative positional relationship changes between the first and second body parts. Rather than processing all raw sensor data, the system isolates and calculates only the relevant features (position, velocity, acceleration relationships) that directly contribute to distinguishing between joystick and touchpad actions, reducing unnecessary computational overhead
Data Source
AI summary
A recognition device includes an acquisition unit and a processor. The acquisition unit acquires first and second informations. The first information relates to a first signal corresponding to a state of a first portion of a body performing an action. The first signal is generated by a first element mounted to the first portion. The second information relates to a second signal corresponding to a state of a second portion. The second signal is generated by a second element mounted to the second portion. A relative positional relationship between the first portion and the second portion changes according to the action. The processor calculates a first feature based on the first information and a second feature based on the second information. The processor recognizes a type of the action based on a change of the first feature and a change of the second feature.