Wearable Force Sensor Surface Classification for Accurate Gait Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wearable devices that measure aspects of a person's movement, such as bodyweight and gait, face errors due to varying surfaces, which are not accurately accounted for by computerized models, especially in diverse terrains like hardpack, asphalt, sand, and grass.
Innovation Solution
A system with a wearable device equipped with force sensors and a computing device that uses AI to classify surface types by training on force sensor data pairs, including input data from the sensors and ground-truth labels, allowing for real-time prediction of surface types during use.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computerized models are used to estimate movement aspects, then the estimation process is simple and fast, but the accuracy deteriorates when surfaces vary significantly from model assumptions
Solution Approach 1:
The patent introduces surface type classification as an intermediary step between the force sensor data collection and the movement estimation. The system first classifies the surface type using force sensor data, then uses this classification to select appropriate movement models, thereby mediating between the physical measurement and the computational estimation to improve accuracy without significantly reducing speed
Solution Approach 2:
The system changes the parameter used for model selection from a fixed assumption to a dynamically determined surface type classification. By determining the surface type as a variable parameter and using it to select different movement models, the system adapts to varying terrain conditions, improving measurement accuracy while maintaining estimation efficiency
2Measurement precision
If surface type classification is added to the measurement system, then measurement accuracy improves, but device complexity increases
Solution Approach 1:
The force sensors in the wearable device perform multiple functions: they measure force data for movement estimation and simultaneously provide data for surface type classification. By making the force sensors multi-functional, the system achieves improved measurement accuracy without adding separate dedicated sensors, thereby limiting the increase in device complexity
Solution Approach 2:
The system performs surface type classification in advance using force sensor data before using this information to guide movement estimation. This preliminary action allows the system to prepare appropriate models ahead of time, improving accuracy while keeping the overall system architecture manageable by breaking down the complex task into sequential steps
Data Source
AI summary
Examples are disclosed that relate to methods and systems for classifying a surface type. One example provides a system comprising a wearable device comprising at least one force sensor, and a computing device having a processor and associated memory storing instructions executable by the processor. The instructions are executable by the processor to, during a training phase, receive training data including a plurality of training data pairs. Each training data pair includes force sensor training data received from the at least one force sensor, or from a simulation or observation, and a label indicating at least one of a plurality of defined surface types. An AI model is trained to predict a classified surface type based on run-time force sensor data. The run-time force sensor data is input into the trained AI model to thereby cause the AI model to output a predicted classification of a run-time surface type.