Adaptive UKF Based Parameter Estimation of Compliant Man-Machine Dynamic Model for Lower Limb Rehabilitation Robot

Abstract

For lower limb rehabilitation robot, the precise man-machine dynamics model is the basis for effective and reliable rehabilitation training. The traditional man-machine system dynamic model is a rigid body model with simplified link structure, which is difficult to realize man-machine synchronization movement without any interference, As well as, the safety of human motion cannot be guaranteed. To solve this problem, An adaptive UKF based parameter estimation of compliant man-machine dynamic model for lower limb rehabilitation robot is put forward. Firstly, according to the compliance of the human body movement, the human dynamic model described in the structure of “connecting rod parallel viscoelastic unit” is established based on the mechanism of human lower limb movement. In addition, the dynamic model of lower limb rehabilitation robot is established. Then, on this basis, the man-machine combined dynamics model is established according to the interaction relationship between human and the rehabilitation robot. Next, the parameters of man-machine dynamic model, which reflect the compliance of human movement, are estimated online by adaptive UKF algorithm. Finally, the compliant man-machine dynamic model is simulated and analyzed. The simulation results show that the adaptive UKF based parameter estimation of compliant man-machine dynamic model is scientific and reasonable.