2020/04 Assist-as-needed Impedance Control Strategy for a Wearable Ankle Robotic Orthosis


“The use of brain-machine interfaces in people who have suffered a stroke can help in their rehabilitation process by cognitively involving the patient. Such interfaces translate brain waves into commands in order to control a mechanical assisted movement device. However, the control of these devices should be more robust and have high precision. This work studies whether algorithms based on Stockwell or Hilbert-Huang transforms can improve the control of these devices by increasing their accuracy, and whether it is advisable to carry out a per-subject and per-electrode configuration customisation. The analysis of five volunteers also shows that it is not possible to detect motor intention from motor event-related desynchronisation/synchronisation with sufficient robustness from pre-movement data alone. Therefore, it is necessary to extend the analysis time to two seconds after the onset of the movement.”