Training with an Industrial Robot: Current Directions to Personalize Neuromuscular Training

Neuromuscular training of the leg extensors is essential in rehabilitation from and prevention of musculoskeletal disorders. To reduce the risk of injury, the forces acting on the musculoskeletal system, the speed of movement, and the range of motion during such training must be continuously monitored and controlled. This talk will explain how an industrial robot and a force plate mounted on its end-effector can be used to overcome the limitations of current training devices. In particular, the presentation will address the application of musculoskeletal models in combination with motion capture to estimate the musculoskeletal load (e.g., joint moments, muscle forces) during exercise. On the one hand, these models must be complex enough to represent the essential characteristics of the person, and, on the other, they must be fast enough to respond to fluctuating loads. Two leg extension training modalities, namely isokinetic (constant speed) and isotonic (constant weight), were designed and implemented. Preliminary simulative and experimental studies have shown that the external adduction moment of the knee, a surrogate measure of medial tibiofemoral contact force, can be minimized using robotics, thereby improving exercise safety for patients with, for example, knee osteoarthritis.