Machine-Learned Mobility Augmentation for Personalized Actuation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobility aids such as crutches, canes, wheelchairs, exoskeletons, and functional electrical stimulation (FES) are not personalized to the user and fail to optimize mobility augmentation based on available information, limiting their effectiveness for individuals with disabilities.
Innovation Solution
A mobility augmentation system using machine learning to monitor muscle electroactivity, kinematics, and kinetics to predict intended movements, applying personalized actuation strategies and optimizing them through real-time feedback loops.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional mobility aids (crutches, canes, wheelchairs, exoskeletons, FES) are used to assist mobility, then basic mobility support is provided, but the system cannot personalize or optimize mobility augmentation based on user-specific movement data
Solution Approach 1:
The system collects and stores movement data from multiple users before actual use. This preliminary data collection enables the system to pre-process and analyze user-specific movement patterns, allowing personalized mobility augmentation to be activated when the user actually needs it, rather than requiring complex real-time analysis from scratch
Solution Approach 2:
A server acts as an intermediary between users and the mobility augmentation system. The server receives movement data from users, processes it to identify user-specific movement patterns, and generates personalized mobility augmentation parameters. This intermediary approach distributes computational complexity away from the user device, enabling personalization without overwhelming local system complexity
2Measurement precision
If machine learning models are trained on generalized movement data from multiple users, then movement prediction accuracy is improved, but the system loses user-specific customization
Solution Approach 1:
The system merges generalized movement data from multiple users with user-specific movement data. The server combines data from many users to establish baseline movement patterns, then overlays individual user data to capture personal characteristics. This combination enables the system to leverage both the statistical power of large datasets and the specificity of individual user patterns
Solution Approach 2:
The system applies different levels of data aggregation to different aspects of movement prediction. Generalized patterns from multiple users are used for baseline movement classification, while user-specific data is applied to refine predictions for individual characteristics. This local quality approach allows the system to use appropriate data granularity for different prediction tasks
3Reliability
If the system continuously monitors and re-trains machine learning models with user feedback, then mobility augmentation optimization is improved, but computational resources and processing time increase
Solution Approach 1:
The system implements periodic re-training of machine learning models rather than continuous re-training. The server schedules re-training operations at intervals, using accumulated movement data to periodically update user-specific movement patterns. This periodic approach maintains model accuracy while avoiding the computational overhead of continuous re-training
Solution Approach 2:
The system performs preliminary processing of movement data as it is collected, organizing and preprocessing data in preparation for future model re-training. This preliminary action reduces the computational burden during actual re-training events, as data is already organized and ready for analysis when re-training occurs
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables personalized and optimized mobility augmentation by predicting intended movements and adjusting actuation strategies to minimize differences between intended and actual movements, improving user independence and mobility.
Implementation Method 1
The system monitors the user for various data associated with movement such as muscle electroactivity (i.e., muscle firing)
Data Source
AI summary
A mobility augmentation system monitors a user's motor intent data and augments the user's mobility based on the monitored motor intent data. A machine-learned model is trained to identify an intended movement based on the monitored motor intent data. The machine-learned model may be trained based on generalized or specific motor intent data (e.g., user-specific motor intent data). A machine-learned model initially trained on generalized motor intent data may be re-trained on user-specific motor intent data such that the machine-learned model is optimized to the movements of the user. The system uses the machine-learned model to identify a difference between the user's monitored movement and target movement signals. Based on the identified difference, the system determines actuation signals to augment the user's movement. The actuation signals determined can be an adjustment to a currently applied actuation such that the system optimizes the actuation strategy during application.