Roboterassistierte Rehabilitation und Muskelaufbautraining
Aachen (2020) [Dissertation / PhD Thesis]
Page(s): 1 Online-Ressource (xiii, 148 Seiten) : Illustrationen, Diagramme
Mobility is an integral component for an independent life even in an advanced age. Regular exercise is crucial for both prevention and rehabilitation of disorders, which can result in impaired mobility. Robot-assisted systems can help to provide patients in neurorehabilitation sufficient independent exercise time and support people in muscular strength training in avoiding bad postures, which might result in damages to the musculoskeletal system. The focus of this work is on the use of industrial robots for neurorehabilitation of the upper extremity and for neuromuscular strength training of the lower extremity. A simulative method, that compares the maximum permissible axes loads with the loads expected in the application is used to evaluate and select a suitable robot. The simulation uses a rigid body model of the robot. A system based on a KUKA LBR IV is used to improve a robotic system for neurorehabilitation. The resulting system allows patients to exercise a previously recorded movement with the robot, independently. The system continuously evaluates the patient's arm posture and detects compensatory movements. This detection is used to assess the patient's current situation with the measured forces at the end effector and to react appropriately. A system based on a KUKA KR270 industrial robot has been developed for neuromuscular strength training, which can be used for both isokinematic and isotonic exercise along arbitrary individual trajectories. In two experiments on a dynamic and static leg press, the position and orientation of the foot plate was identified as a manipulated variable in order to control loads on the musculoskeletal system. With motion detection as well as a rigid body model of the leg, the measured force at the end effector can be used to determine the joint loadings during exercise. This is the basis for a management of loadings in neuromuscular strength training, which is now possible with the robot-assisted training system developed in this thesis.