Neural-Network Based Iterative Learning Control of a Hybrid Exoskeleton with an MPC Allocation Strategy

In this paper, a novel neural network based iterative learning controller for a hybrid exoskeleton is presented. The control allocation between functional electrical stimulation and knee electric motors uses a model predictive control strategy. Further to address modeling uncertainties, the controller identifies the system dynamics and input gain matrix with neural networks in an iterative fashion. Virtual constraints are employed so that the system can use a time invariant manifold to determine desired joint angles. Simulation results show that the controller stabilizes the hybrid system for sitting to standing and standing to sitting scenarios.