A decentralized adaptive controller design for Lower Extremity Rehabilitation Robot

Rehabilitation is the major therapy in stroke and spinal cord injured individuals. Robot-aided treadmill training has been applied for several years to assist, enhance and evaluate neurological and orthopedic rehabilitation. This paper presents a compliant patient cooperative control approach for Lower Extremity Rehabilitation Robot (LERR) with automatic gait adaption. Firstly, based on nonlinear inverbility decoupling theory the system is decoupled into some independent second-order integral systems. Then based on the decoupled sub-systems, the admittance and adaptive control methods are further investigated for the gait trajectory planning and tracking. Finally, the performance of the proposed controller is verified in the MATLAB-Adams co-simulation environment.