Wearable electromyography sensor array using conductive cloth electrodes for human-robot interactions
A sleeve worn on an arm allows detection of gestures by an array of sensors. Electromyography, inertial, and magnetic field sensors provide data that is processed to categorize gestures and translate the gestures into commands for robotic systems. Machine learning allows training of gestures to increase accuracy of detection for different users.
1. A device comprising:
a plurality of electrodes attached on an inner surface of an elastic sleeve and configured to detect electric signals on a skin;
a plurality of electromyography sensors on an outer surface of the elastic sleeve, the plurality of electrodes being electrically connected to the plurality of electromyography sensors through the elastic sleeve; and
at least one inertial measurement unit attached to the elastic sleeve and configured to detect an orientation of the elastic sleeve,
wherein each electrode of the plurality of electrodes comprises conductive cloth wrapped around a foam or rubber insert, and a snap-on attachment configured to mechanically and electrically connect to a corresponding electromyography sensor through a corresponding opening in the elastic sleeve.
2. The device of claim 1 , further comprising a processor configured to process data from the plurality of electromyography sensors and from the at least one inertial measurement unit to detect hand and forearm gestures.
3. The device of claim 2 , wherein the processor comprises a gesture recognition software to decode the sensor signals, classify gesture type, and map the result to output commands to be sent to a robot.
4. The device of claim 1 , wherein the at least one inertial measurement unit comprises a 3-axis gyroscope, 3-axis accelerometer, and 3-axis magnetometer.
5. The device of claim 1 , wherein the at least one inertial measurement unit comprises a plurality of inertial measurement units.
6. The device of claim 1 , wherein the conductive cloth comprises nylon and silver.
7. A device comprising:
a plurality of electrodes attached on an inner surface of an elastic sleeve and configured to detect electric signals on a skin;
a plurality of electromyography sensors on an outer surface of the elastic sleeve, the plurality of electrodes being electrically connected to the plurality of electromyography sensors through the elastic sleeve; and
at least one inertial measurement unit attached to the elastic sleeve and configured to detect an orientation of the elastic sleeve,
wherein each electrode of the plurality of electrodes comprises conductive cloth attached to one side of an electric gel pad, and a snap-on attachment on an opposite side of the electric gel pad, to mechanically and electrically connect to a corresponding electromyography sensor through a corresponding opening in the elastic sleeve.
8. A method comprising:
providing a device to be worn on a forearm, the device comprising:
a plurality of electrodes attached on an inner surface of an elastic sleeve,
a plurality of electromyography sensors on an outer surface of the elastic sleeve, the plurality of electrodes being electrically connected to the plurality of electromyography sensors through the elastic sleeve,
at least one inertial measurement unit attached to the elastic sleeve, and
a processor;
detecting electric signals on a forearm skin by the plurality of electromyography sensors;
detecting an orientation of the forearm by the at least one inertial measurement unit;
categorizing muscle activations in the forearm into corresponding hand and forearm gestures, based on the detected electric signals and orientation; and
issuing commands to a robotic system based on the hand and forearm gestures,
wherein each electrode of the plurality of electrodes comprises conductive cloth wrapped around a foam or rubber insert, and a snap-on attachment to mechanically and electrically connect to a corresponding electromyography sensor through a corresponding opening in the elastic sleeve.
9. The method of claim 8 , further comprising providing a library of hand and forearm gestures corresponding to commands for the robotic system.
10. The method of claim 9 , further comprising training the device to categorize the hand and forearm gestures by instructing a user to perform each hand and forearm gesture in the library for a set number of times and a set period.
11. The method of claim 8 , wherein the conductive cloth comprises nylon and silver.
12. The method of claim 8 , wherein the at least one inertial measurement unit comprises a 3-axis gyroscope, 3-axis accelerometer, and 3-axis magnetometer.
13. The method of claim 10 , wherein the training and categorizing are based on machine learning algorithms.