Automatic versus manual tuning of robot-assisted gait training in people with neurological disorders

This study compares automatic and manual tuning of robotic assistance during gait training for people with neurological disorders. The goal was to see if an algorithm could adjust the robotic assistance as well as an experienced therapist, focusing on various subtasks of walking. Ten participants (six stroke, four spinal cord injury) walked in a robotic gait trainer with both automatically-tuned (AT) and manually-tuned (MT) assistance. Assistance was adjusted for subtasks such as weight shift, foot placement, and limb angle control. The results showed that the automatic tuning was quicker and used lower assistance levels, but both methods were rated similarly by participants for safety, comfort, effect, and amount of assistance.

• 1
  Automatic tuning (AT) reached stable assistance levels faster than manual tuning (MT).
• 2
  Considerable differences were found in the assistance provided per subtask by the automatic and manual approaches. The amount of assistance was more often higher for the manual approach.
• 3
  Despite higher assistance levels in the manual approach, the largest deviations from the reference trajectories were found for the manual algorithm.