Name: PEDRO HENRIQUE FABRIZ ULHOA
Publication date: 26/06/2026
Examining board:
Name |
Role |
|---|---|
| CAMILO ARTURO RODRIGUEZ DIAZ | Examinador Interno |
| PERCY NOHAMA | Examinador Externo |
| RAFHAEL MILANEZI DE ANDRADE | Coorientador |
| TEODIANO FREIRE BASTOS FILHO | Presidente |
Summary: Robotic prostheses have emerged as an alternative to passive prostheses, which are
limited to energy dissipation during locomotion and may therefore compromise amputee
biomechanics. However, most controllers implemented in robotic prostheses are non
volitional. In other words, they are not integrated with the user’s nervous system, preventing
amputees from intentionally controlling the robotic leg. This research project aims to
contribute to the development of myoelectric controllers for lower-limb robotic prostheses
using sEMG as volitional input. To accomplish this goal, a controller based on EMG
signals captured from muscles responsible for the flexion and extension of the knee joint
was implemented on the onLeg, a robotic prosthesis developed at the Robotics and
Biomechanics Laboratory of UFES. The controller is based on the muscular co-contraction
exhibited by the antagonist muscles of the thigh. This muscular co-contraction is used
to modulate the stiffness and speed of the robotic knee prosthesis depending on the
activity or locomotion phase being performed. In this research, EMG signal data from
transfemoral amputees and healthy individuals were collected for a preliminary evaluation
of the muscular co-contraction of both groups. Subsequently, the proposed controller was
used by two transfemoral amputees while performing walking on level ground, standing
up from a chair, and avoiding obstacles. With the assistance of the robotic prosthesis
controlled by EMG signals, the amputee volunteers performed the sit-to-stand movement
with a time reduction of approximately 32% e 37%. Walking on level ground was also
possible with the proposed controller. However, the knee ranges of motion during gait were
lower than those of healthy individuals and differed by 53° between the two volunteers.
The amputee individuals were able to avoid obstacles of 7, 10, 13, and 16 cm in height,
modulating the foot clearance through the combined control of the biological hip joint and
the robotic knee joint under myoelectric control. The results suggest that the implemented
controller allows the amputee to have greater control over the prosthesis, influencing its
state according to external demand, a characteristic that proves advantageous in various
scenarios of activities of daily living.
