Recognition method of human walking speed intention from surface electromyogram signals of plantar flexor and walking speed control method of a lower-limb exoskeleton robot
A walk assist robot for lower body walking of a walking trainee, including a joint angle signal measurement unit disposed on a joint of the walking trainee, an electromyogram (EMG) signal measurement unit disposed on a muscle related to ankle joint extension of the walking trainee, a plantar pressure signal measurement unit disposed on a sole of the walking trainee, and a control unit to recognize signals measured from the joint angle signal measurement unit, the EMG signal measurement unit and the plantar pressure signal measurement unit and process the signals to recognize a walking speed intention of the walking trainee, wherein the control unit controls a walking speed of the walk assist robot from the walking speed intention of the walking trainee.
1. A walk assist robot for lower body walking of a walking trainee, comprising:
a joint angle sensor adapted to be disposed on a joint of the walking trainee;
an electromyogram (EMG) signal sensor adapted to be disposed on a muscle related to ankle joint extension of the walking trainee;
a plantar pressure sensor adapted to be disposed on a sole of the walking trainee; and
a controller configured to recognize measured signals from the joint angle sensor, the EMG signal sensor, and the plantar pressure sensor, and process the recognized signals to recognize a walking speed intention of the walking trainee,
wherein the controller controls a walking speed of the walk assist robot from the recognized walking speed intention of the walking trainee.
2. The walk assist robot according to claim 1 , wherein the controller further comprises:
a band-pass filter to process a joint angle signal from the joint angle sensor and plantar pressure signal from the plantar pressure sensor;
a waveform length calculator to calculate a waveform length from an EMG signal from the EMG signal sensor; and
a stance phase detector to detect a stance phase range of walking of the walking trainee.
3. The walk assist robot according to claim 2 , wherein the walking speed intention of the walking trainee is repetitively recognized each point in time at which the stance phase of walking ends.
4. The walk assist robot according to claim 1 , wherein the EMG signal sensor is adapted to be arranged at a position of at least one of soleus, gastrocnemius, and peroneus longus muscles in the muscle related to ankle joint extension of the walking trainee.
5. The walk assist robot according to claim 1 , wherein the plantar pressure unit sensor is adapted to be arranged at a position of at least one of a medial forefoot, a lateral forefoot and a rearfoot of left and right soles of the walking trainee.
6. The walk assist robot according to claim 1 , wherein the joint angle sensor is adapted to be arranged at left and right knee joints or angle joints of the walking trainee.
7. The walk assist robot according to claim 2 , wherein the controller configured to estimate a walking speed intention in a swing phase during a gait cycle of the walking trainee, using a maximum value of the waveform length of the EMG signal in the stance phase during the gait cycle of the walking trainee.
8. A processor implemented method of controlling a walking speed of a walk assist robot using the walk assist robot, the method comprising:
measuring a joint angle signal from a joint angle sensor;
measuring an electromyogram (EMG) signal from an EMG signal sensor;
measuring a plantar pressure signal from a plantar pressure sensor;
performing the respective signal-processing of at least one of the joint angle signal from the joint angle sensor, the EMG signal from the EMG signal sensor, and the plantar pressure signal from plantar pressure sensor;
calculating a waveform length from the signal-processed EMG signal;
detecting a stance phase from the signal-processed joint angle signal and the signal-processed plantar pressure signal;
recognizing a walking speed intention from the waveform length and the stance phase;
generating a walking trajectory of the walk assist robot from the walking speed intention; and
controlling a walking speed of the walk assist robot from the walking trajectory of the walk assist robot.
9. The method according to claim 8 , wherein the joint angle sensor adapted to be disposed on a joint of the walking trainee;
the electromyogram (EMG) signal sensor adapted to be disposed on a muscle related to ankle joint extension of the walking trainee; and
the plantar pressure sensor adapted to be disposed on a sole of the walking trainee.
10. A walk assist robot system for lower body walking of a walking trainee, comprising:
a joint angle sensor adapted, in the walk assist robot system, to be disposed on a joint of the walking trainee;
an electromyogram (EMG) signal sensor adapted, in the walk assist robot system, to be disposed on a muscle related to ankle joint extension of the walking trainee;
a plantar pressure sensor adapted, in the walk assist robot system, to be disposed on a sole of the walking trainee; and
a controller configured to recognize measured signals from the joint angle sensor, the EMG signal sensor, and the plantar pressure sensor, and process the recognized signals to recognize a walking speed intention of the walking trainee,
wherein the controller controls a walking speed of the walk assist robot from the walking speed intention of the walking trainee.
11. The system according to claim 10 , further comprising the walk assist robot.
12. The system according to claim 10 , wherein controller further comprises:
a band-pass filter to process a joint angle signal from the joint angle sensor and a plantar pressure signal from the plantar pressure sensor;
a waveform length calculator to calculate a waveform length from an EMG signal from the EMG signal sensor; and
a stance phase detector to detect a stance phase range of walking of the walking trainee.
13. The system according to claim 10 , wherein the walking speed intention of the walking trainee is repetitively recognized each point in time at which the stance phase of walking ends.
14. The system according to claim 10 , wherein the EMG signal sensor is adapted, in the walk assist robot system, to be arranged at a position of at least one of soleus, gastrocnemius, and peroneus longus muscles in the muscle related to ankle joint extension of the walking trainee.
15. The system according to claim 10 , wherein the plantar pressure sensor is adapted, in the walk assist robot system, to be arranged at a position of at least one of a medial forefoot, a lateral forefoot and a rearfoot of left and right soles of the walking trainee.
16. The system according to claim 10 , wherein the joint angle sensor is adapted, in the walk assist system, to be arranged at left and right knee joints or angle joints of the walking trainee.
17. The walk assist robot according to claim 12 , wherein the controller configured to estimate a walking speed intention in a swing phase during a gait cycle of the walking trainee, using a maximum value of the waveform length of the EMG signal in the stance phase during the gait cycle of the walking trainee.