Enhanced activated exoskeleton system
An enhanced exoskeleton system is disclosed comprising an exoskeleton, a base layer comprising at least one sensor, an exoskeleton actuator configured to actuate the exoskeleton, a control subsystem comprising one or more processors, and memory elements including instructions that, when executed, cause the processors to perform operations comprising: receiving sensor data from the sensor, determining a future movement intent of a user of the exoskeleton, determining a command for an exoskeleton actuator based on the future movement intent, and communicating the command to the exoskeleton actuator whereby the exoskeleton is actuated by the exoskeleton actuator. In some embodiments, the sensor data comprises an anticipatory data value from an anticipatory sensor. In some embodiments, the sensor data comprises an anticipatory data value from an electromyography (EMG) sensor.
1. A control subsystem for controlling an actuator of an exoskeleton based on a future movement intent of a user, the control subsystem comprising:
a control database storing a historical movement library;
a current intent classifier configured to determine a current movement intent of the user based on a sensor data and the historical movement library;
a future intent classifier configured to determine a future movement intent of the user based on the current movement intent of the user;
a motor control module configured to determine a command to be communicated to an actuator of the exoskeleton; and
the command is based on the future movement intent of the user.
2. The control subsystem of claim 1 wherein the sensor data comprises an anticipatory data value from an anticipatory sensor.
3. The control subsystem of claim 1 wherein the sensor data comprises an anticipatory data value from an electromyography (EMG) sensor.
4. The control subsystem of claim 1 wherein:
the sensor data comprises an anticipatory data value from an electromyography (EMG) sensor; and
the anticipatory data value representing a muscle movement of the user that has not occurred.
5. The control subsystem of claim 1 wherein the current intent classifier is configured to determine the current movement intent of the user according to a probabilistic pattern recognition technique.
6. The control subsystem of claim 1 wherein the future intent classifier is configured to determine the future movement intent of the user according to a probabilistic pattern recognition technique.
7. The control subsystem of claim 1 wherein the current intent classifier is configured to determine the current movement intent of the user according to a K-Nearest Neighbor algorithm.
8. The control subsystem of claim 1 wherein the future intent classifier is configured to determine the future movement intent of the user according to a K-Nearest Neighbor algorithm.
9. The control subsystem of claim 1 further comprising a current task classifier configured to determine a current movement task of the user according to a probabilistic pattern recognition technique.
10. The control subsystem of claim 1 further comprising a current task classifier configured to determine a current movement task of the user according to a K-Nearest Neighbor algorithm.
11. An enhanced exoskeleton system comprising the control subsystem of claim 1 , the enhanced exoskeleton system further comprising:
a sensor-embedded textile base layer configured to sense the sensor data and communicate the sensor data to the control subsystem; an active exoskeleton in communication with the control subsystem and configured to be controlled by the actuator; and
the textile base layer comprises an electromyography (EMG) sensor configured to communicate an anticipatory data value representing a muscle movement yet to occur.