Haptic Vibration Control Using Body Condition Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing haptic devices lack the ability to selectively present appropriate tactile/force senses to users undergoing rehabilitation for movement disorders, as the type and intensity of these sensations are not tailored to the individual's specific body condition.
Innovation Solution
A control unit for a haptic device that includes a storage and an execution unit, which stores vibration mode data and model data for a learning model. This unit obtains body condition variables from the user, selects a specific vibration mode based on the learning model's output, and drives the vibrator to provide a suitable tactile/force sense for rehabilitation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a haptic device uses fixed vibration patterns for rehabilitation, then the device structure is simple and easy to operate, but it cannot adapt to different user body conditions and rehabilitation needs
Solution Approach 1:
The system performs self-diagnosis by automatically acquiring body condition variables (temperature, pulse, respiration) through sensors and using machine learning to autonomously select appropriate vibration modes without requiring manual input from users or therapists, enabling the device to adapt to different users independently
Solution Approach 2:
The system continuously monitors body condition variables and uses machine learning models to adjust vibration patterns in real-time based on the user's physiological state, creating a closed-loop feedback system that adapts the haptic output to the user's current condition
2Adaptability or versatility
If the haptic device presents multiple vibration modes for different rehabilitation needs, then it can better suit individual users, but the control system becomes more complex
Solution Approach 1:
The system automatically acquires body condition variables through integrated sensors and uses machine learning to autonomously select the most appropriate vibration mode from multiple available patterns, eliminating the need for users or therapists to manually select modes and simplifying operation while maintaining high adaptability
3Adaptability or versatility
If the haptic device uses standardized vibration patterns, then the device is easier to manufacture and operate, but it cannot provide tailored sensory feedback for individual rehabilitation needs
Solution Approach 1:
The system changes the parameters of vibration patterns (frequency, amplitude, duration, pattern type) based on machine learning analysis of body condition variables, allowing the same physical device to generate customized tactile feedback by dynamically adjusting vibration parameters rather than requiring multiple fixed-pattern devices
Data Source
AI summary
A control unit includes a CPU as an execution unit and a storage. The storage stores vibration mode data indicating multiple vibration modes used for rehabilitation. The storage stores model data which determines a learning model. A body condition variable indicating body information of a user is input into the learning model and a mode variable is output from the learning model. The mode variable indicates a pattern of a vibration mode that represents a tactile/force sense to output from a haptic device. The model data is learned data obtained by machine learning. In obtaining processing, the CPU obtains multiple body condition variables of the user. In mode selection processing, the CPU selects a specific mode, based on mode variables output by using the multiple body condition variables obtained in the mode obtaining processing as input variables. In driving processing, the CPU drives a vibrator by using the specific mode.