SYSTEM AND METHOD FOR CREATING CUSTOM FOOTWEAR
A system, method and computer program product for creating a custom footwear article design. A plurality of sensor readings are obtained from a corresponding plurality of sensors provided in a wearable device while the wearable device is worn by a user while also wearing a test footwear article. At least one customization criterion can be identified for the user. A plurality of biomechanical contributing factor values can be determined for the user based on the sensor readings. A cost value can be determined based on differences between the biomechanical contributing factor values and a plurality of preferred biomechanical contributing factor values. A footwear article design is optimized by adjusting at least one characteristic of the test footwear article to minimize the cost value. The custom footwear article design can be identified as the footwear article design corresponding to a minimum cost value.
1 . A method for creating a custom footwear article design, the method comprising:
obtaining a plurality of sensor readings from a corresponding plurality of sensors provided in a wearable device, and while the wearable device is worn by a user while also wearing a test footwear article;
identifying at least one customization criterion for the user;
determining a plurality of biomechanical contributing factor values for the user based on the sensor readings;
determining a cost value based on at least one difference between the plurality of biomechanical contributing factor values and a plurality of preferred biomechanical contributing factor values;
optimizing a footwear article design by adjusting at least one characteristic of the test footwear article to minimize the cost value; and
identifying the custom footwear article design as the footwear article design corresponding to a minimum cost value.
2 . The method of claim 1 , further comprising manufacturing the custom footwear article using the custom footwear article design.
3 . The method of claim 1 , further comprising:
identifying a preferred existing footwear article similar to the custom footwear article; and
outputting an indication of the preferred existing footwear article.
4 . The method of claim 1 , wherein
each biomechanical contributing factor value has an associated weight representing a level of contribution of a corresponding biomechanical contributing factor to the at least one customization criterion;
each difference of the at least one difference is used to determine a factor-specific error value; and
the cost value is determined by weighing each factor-specific error value according to the associated weight for the corresponding biomechanical contributing factor.
5 . The method of claim 1 , wherein the design of the footwear article is optimized iteratively using finite element analysis of a model of the design of the footwear article.
6 . The method of claim 1 , wherein the plurality of sensors comprises one or more of a force-sensing element and an inertial measurement unit.
7 . The method of claim 1 , wherein the biomechanical contributing factor values comprise at least one of: a pressure value, a ground reaction force value, a center of pressure value, a foot contact event value, a stride time value, a ground contact time value, a swing time value, a rate of pronation value, a rate of force development value, a foot strike index value, a foot orientation value, a mid-stance value, a stride length value, a stride velocity value, a joint force value, a joint moment value, or a power value.
8 . The method of claim 1 , wherein the at least one characteristic comprises at least one of a material hardness, a material density, a material stiffness, a material texture, a footwear shape, footwear contouring, a footwear size, a footwear component number, a footwear component size, a footwear component location, or a toebox volume.
9 . The method of claim 1 , wherein the at least one customization criterion comprises a specified customization metric, and the specified customization metric is one of an injury risk metric, a performance level metric, and a discomfort level metric.
10 . The method of claim 9 , wherein
the plurality of preferred biomechanical contributing factor values are defined based on an analysis of a machine learning model trained to determine a value of the specified customization metric, wherein the analysis of the machine learning model identifies the preferred biomechanical contributing factor values as those biomechanical contributing factor values that contribute to a desired metric value of the specified customization metric;
the machine learning model is trained using training data that includes training sensor readings obtained from a plurality of training users performing one or more specified activities, wherein the plurality of users are associated with different metric values of the specified customization metric; and
the training data comprises training user anthropometric data and training user wellness data for the plurality of training users.
11 . A system for creating a custom footwear article design comprising:
a plurality of sensors mountable to a user, wherein the plurality of sensors is configured to obtain a plurality of sensor readings from the user while the user is also wearing a test footwear article;
a memory configured to store the plurality of sensor readings and a plurality of preferred biomechanical contributing factor values; and
one or more processors configured to:
identify at least one customization criterion for the user;
determine a plurality of biomechanical contributing factor values for the user based on the sensor readings;
determine a cost value based on at least one difference between the plurality of biomechanical contributing factor values and the plurality of preferred biomechanical contributing factor values;
optimize a footwear article design by adjusting at least one characteristic of the test footwear article to minimize the cost value; and
identify the custom footwear article design as the footwear article design corresponding to a minimum cost value.
12 . The system of claim 11 , wherein the one or more processors is configured to:
identify a preferred existing footwear article similar to the custom footwear article; and
output an indication of the preferred existing footwear article.
13 . The system of claim 11 , wherein the one or more processors is configured to:
associate each biomechanical contributing factor value with an associated weight representing a level of contribution of a corresponding biomechanical contributing factor to the at least one customization criterion;
use each difference of the at least one difference to determine a factor-specific error value; and
determine the cost value by weighing each factor-specific error value according to the associated weight for the corresponding biomechanical contributing factor.
14 . The system of claim 11 , wherein the one or more processors is configured to optimize the design of the footwear article iteratively using finite element analysis of a model of the design of the footwear article.
15 . The system of claim 11 , wherein the plurality of sensors comprises one or more of a force-sensing element and an inertial measurement unit.
16 . The system of claim 11 , wherein the biomechanical contributing factor values comprise at least one of: a pressure value, a ground reaction force value, a center of pressure value, a foot contact event value, a stride time value, a ground contact time value, a swing time value, a rate of pronation value, a rate of force development value, a foot strike index value, a foot orientation value, a mid-stance value, a stride length value, a stride velocity value, a joint force value, a joint moment value, or a power value.
17 . The system of claim 11 , wherein the at least one characteristic comprises at least one of a material hardness, a material density, a material stiffness, a material texture, a footwear shape, footwear contouring, a footwear size, a footwear component number, a footwear component size, a footwear component location, or a toebox volume.
18 . The system of claim 11 , wherein the one or more processors is configured to identify the at least one customization criterion to include a specified customization metric, and the specified customization metric is one of an injury risk metric, a performance level metric, and a discomfort level metric.
19 . The system of claim 18 , wherein
the plurality of preferred biomechanical contributing factor values are defined based on an analysis of a machine learning model trained to determine a value of the specified customization metric, wherein the analysis of the machine learning model identifies the preferred biomechanical contributing factor values as those biomechanical contributing factor values that contribute to a desired metric value of the specified customization metric;
the machine learning model is trained using training data that includes training sensor readings obtained from a plurality of training users performing one or more specified activities, wherein the plurality of users are associated with different metric values of the specified customization metric; and
the training data comprises training user anthropometric data and training user wellness data for the plurality of training users.
20 . The system of claim 11 , wherein the wearable device is footwear, and the footwear is an insole.