Personalized Robotic Training on a Planar Reaching Task

Abstract

Based on recent advancements in neuro-adaptive control, we evaluated a novel iterative algorithm for generating customized training forces. The objective was to study the motor adaptation required to compensate for robot-generated perturbations during an upper-limb reaching task. We hypothesized that the adaptation process would induce changes in the brain's feed-forward command, dynamically reshaping the neuromuscular system to refine movement patterns and result in trajectory alterations. The results indicate that after a training period with these robot-generated forces, trajectories undergo modifications due to internal model adaptation, leading to improved performance measured in terms of position error. This experiment explores motor learning in different directions and compares two conditions: 1) curl force field, a velocitydependent perturbation applied to deviate trajectories, and 2) Error Field force, a customized force specifically designed to address and correct trajectory errors. The findings highlight the effectiveness of the Error Field force in enhancing motor learning and show the potential of customized robotic perturbation forces for improving motor performance in personal activities and advancing neurorehabilitation techniques.