Classifying a surface type using force sensor data

At a computing device, a method for classifying a surface type, the method comprising:

• during a training phase,
• receiving training data including a plurality of training data pairs, each training data pair including an input vector and a matched ground-truth label,
• wherein the input vector includes force sensor training data, wherein the force sensor training data is received from at least one force sensor of a wearable device, the at least one force sensor located between a user's foot and a surface, the user traversing that surface during a training session, or from a simulation or observation of one or more other force sensors, and
• wherein the matched ground-truth label indicates at least one of a plurality of defined surface types;
• and
• training an artificial intelligence (AI) model using the plurality of training data pairs to predict a classified surface type based on run-time force sensor data;
• and
• during a run-time phase,
• monitoring the run-time force sensor data from the at least one force sensor as the user is traversing a run-time surface type, wherein the run-time surface type is one of the plurality of defined surface types;
• and
• inputting the run-time force sensor data into the trained AI model to thereby cause the AI model to output a predicted classification of the run-time surface type.

A system, comprising:

• a wearable device comprising at least one force sensor located between a user's foot and a surface;
• and
• a computing device having a processor and associated memory storing instructions executable by the processor to:
• during a training phase,
• receive training data including a plurality of surface training data pairs and a plurality of load training data pairs,
• each surface training data pair including an input vector and a matched ground-truth label,
• wherein the input vector includes force sensor training data, wherein the force sensor training data is received from the at least one force sensor as the user is traversing the surface during a training session, or from a simulation or observation of one or more other force sensors, and
• wherein the matched ground-truth label indicates at least one of a plurality of defined surface types;
• and
• each load training data pair including,
• as input, the force sensor training data, wherein the force sensor training data is received from the at least one force sensor while the user is bearing at least one load of a plurality of defined loads, while simulating the at least one load, or while observing one or more other force sensors bearing the at least one load, and
• as ground-truth output, a ground-truth load classification of the at least one load;
• train a surface classification artificial intelligence (AI) model using the plurality of surface training data pairs to predict a classified surface type based on run-time force sensor data;
• and
• train a load-determining AI model using the plurality of load training data pairs to predict a run-time load classification based on the run-time force sensor data;
• and
• during a run-time phase,
• monitor the run-time force sensor data from the at least one force sensor as the user is traversing a run-time surface type, wherein the run-time surface type is one of the plurality of defined surface types;
• input the run-time force sensor data into the trained surface classification AI model to thereby cause the surface classification AI model to output a predicted classification of the run-time surface type;
• and
• input the run-time force sensor data into the trained load-determining AI model to thereby cause the load-determining AI model to output a predicted classification of a run-time load.