Due to the growing aging society and the increasing number of people who suffer from lower limb
disorders worldwide, the need for robot-assisted gait training devices has become increasingly
urgent. These systems are essential for supporting clinicians in rehabilitation and enhancing
functional recovery in both upper and lower extremities. However, the significant variability in
users' physical characteristics—such as body mass, limb length, and inertia—introduces dynamic
uncertainties that challenge conventional control strategies. To address this, we propose a User-
Adaptive Genetic Algorithm-Tuned Neuro-Fuzzy (GA-Tuned Neuro-Fuzzy) controller integrated
within a Computed Torque Control (CTC) framework. The proposed approach leverages the
learning capability of an Adaptive Neuro-Fuzzy Inference System (ANFIS) and the optimization
ability of a Genetic Algorithm (GA) to adaptively tune control parameters based on varying
anthropometric profiles. Simulation results demonstrate that the controller achieves accurate
trajectory tracking, low steady-state error, and robust performance across a wide range of user
conditions and reference inputs. These findings validate the effectiveness of the proposed controller
for personalized and adaptive rehabilitation in lower limb exoskeleton applications.