Determining readiness for an operational task

An edge computing device, comprising:

• a processor;
• and
• a memory storing instructions executable by the processor to:
• during a training phase, for each of a plurality of user training sessions associated with a training task performed by a user, wherein the training task comprises foot travel during the user training session:
• receive location sensor data from a location sensor communicatively coupled to the processor of the edge computing device, wherein the location sensor comprises one or more of an accelerometer or a global positioning system (GPS) sensor;
• use the location sensor data to determine distance of travel and a time elapsed for performance of the training task by the user during the user training session;
• receive a terrain type and a weather condition in which the training task is performed during the user training session, the terrain type being selected from a plurality of predefined terrain types;
• receive biometric data for the user from a biometric sensor, the biometric data comprising a biometric parameter, wherein the biometric sensor comprises one or more of a heart rate monitor, a galvanic skin response (GSR) sensor, a pulse oximeter, or a breath gas analyzer;
• generate a training data pair for the user training session, the training data pair including,
• as input, the distance of travel, the weather condition, the terrain type, and the biometric parameter, and
• as ground truth output, the time elapsed determined using the location sensor data;
• train an artificial intelligence (AI) performance model that models user performance of the foot travel during the user training session based on the training data pairs for the plurality of the user training sessions;
• and
• during a run-time phase:
• receive operational input data associated with an operational task performed by the user, the operational task comprising foot travel, the operational input data including a target distance of travel, a target weather condition, a target terrain type in which the operational task is performed, and a value for the biometric parameter for the user during the run-time phase;
• based on the operational input data, use the AI performance model to infer a predicted time elapsed for performance of the operational task;
• output the predicted time elapsed for performance of the operational task;
• receive feedback for the predicted time elapsed for performance of the operational task;
• pair the feedback with the operational input data as a feedback training data pair;
• and
• use the feedback training pair to conduct feedback training on the AI performance model.

A system for determining a user's readiness for an operational task, the system comprising:

• one or more location sensors configured to output location sensor data indicating a time elapsed for performance of a training task during a training phase, wherein the one or more location sensors comprise one or more of an accelerometer or a global positioning system (GPS) sensor;
• one or more biometric sensors, wherein the one or more biometric sensors comprise one or more of a heart rate monitor, a galvanic skin response (GSR) sensor, a pulse oximeter, or a breath gas analyzer;
• an edge computing device, comprising,
• a processor communicatively coupled to the one or more location sensors, and
• a memory storing instructions executable by the processor to,
• during the training phase, for each of a plurality of user training sessions associated with the training task, wherein the training task comprises foot travel during the user training session:
• receive the location sensor data from the one or more location sensors;
• use the location sensor data to determine distance of travel and a time elapsed for performance of the training task by the user during the user training session;
• receive a terrain type and a weather condition in which the training task is performed during the user training session, the terrain type being selected from a plurality of predefined terrain types;
• receive biometric data for the user from the one or more biometric sensors, the biometric data comprising at least one biometric parameter;
• generate a training data pair for the user training session, the training data pair including,
• as input, the distance of travel, the weather condition, the terrain type and the at least one biometric parameter, and
• as ground truth output, the time elapsed determined using the location sensor data;
• train an artificial intelligence (AI) performance model that models user performance of the foot travel during the user training session based on the training data pairs for the plurality of the user training sessions, and
• during a run-time phase:
• receive operational input data associated with an operational task performed by the user, the operational task comprising foot travel, the operational input data including a target distance of travel, a target weather condition, a target terrain type in which the operational task is performed, and a value for the biometric parameter for the user during the run-time phase,
• based on the operational input data, use the AI performance model to infer a predicted time elapsed for performance of the operational task,
• output the predicted time elapsed for performance of the operational task;
• receive feedback for the predicted time elapsed for performance of the operational task;
• pair the feedback with the operational input data as a feedback training data pair;
• and
• use the feedback training pair to conduct feedback training on the AI performance model.

A method for determining a user's readiness for an operational task, the method comprising:

• at one or more processors of a wearable edge computing device co-located with the user:
• during a training phase, for each of a plurality of user training sessions associated with a training task performed by the user, wherein the training task comprises foot travel during the user training session:
• receiving location sensor data from a location sensor communicatively coupled to the one or more processors of the wearable edge computing device, wherein the location sensor comprises one or more of an accelerometer or a global positioning system (GPS) sensor;
• using the location sensor data to determine distance of travel and a time elapsed for performance of the training task by the user during the user training session;
• receiving a terrain type and a weather condition in which the training task is performed during the user training session, the terrain type being selected from a plurality of predefined terrain types;
• receiving biometric data for the user from a biometric sensor, the biometric data comprising a biometric parameter, wherein the biometric sensor comprises one or more of a heart rate monitor, a galvanic skin response (GSR) sensor, a pulse oximeter, or a breath gas analyzer;
• generating a training data pair for the user training session, the training data pair including,
• as input, the distance of travel, the weather condition, the terrain type, and the time elapsed for performance of the training task by the user during the training session determined using the location sensor data, and
• as ground-truth output, the biometric parameter;
• training an artificial intelligence (AI) performance model that models user performance of the foot travel during the user training session based on the training data pairs for the plurality of the user training sessions;
• and
• during a run-time phase:
• receiving operational input data associated with an operational task performed by the user, the operational task comprising foot travel, the operational input data including a target distance of travel, a target weather condition, a target terrain type in which the operational task is performed, and a time elapsed for performance of the operational task;
• based on the operational input data, using the AI performance model to infer a predicted value for the biometric parameter for the user during the run-time phase;
• outputting the predicted value for the biometric parameter;
• receiving feedback for the predicted value for the biometric parameter;
• pairing the feedback with the operational input data as a feedback training data pair;
• and
• using the feedback training pair to conduct feedback training on the AI performance model.