Adaptive body weight support controls human activity during robot-aided gait training

Abstract

Current clinical practice of robot-aided gait training is not as effective as expected. Cooperative control strategies aim at improving the effectiveness of robot-aided training by empowering patients to participate more actively. Our group has recently proposed the concept of bio-cooperative control, which explicitely considers the role of the human in the loop, as an extension of these strategies. A supervising controller adapts the cooperative control loops in a way that guarantees appropriate stimuli and prevents undue stress or harm for the patients. In this paper, we implement this concept with an adaptive body weight support algorithm. The algorithm was evaluated with the Lokomat gait rehabilitation robot and the Lokolift body weight support system. Experiments showed that human activity was successfully controlled during Lokomat walking. The desired level of activity was effectively limited when subjects simulated weakness in load bearing. The proposed algorithm may help to train patients with neurological gait impairments in a more engaging and, thus, hopefully more effective way.