Robust Trajectory Free Model Predictive Control of Biped Robots with Adaptive Gait Length

This paper employs nonlinear disturbance observer (NDO) for robust trajectory-free Nonlinear Model Predictive Control (NMPC) of biped robots. The NDO is used to reject the additive disturbances caused by parameter uncertainties, unmodeled dynamics, joints friction, and external slow-varying forces acting on the biped robots. In contrary to the slow-varying disturbances, handling sudden pushing disturbances acting on the biped robots is much more complicated and using the NDO doesn’t guarantee the biped walking stability. In order to reject these kinds of disturbances, the motion controller must be able to make suitable decisions for quick changing of the gait length or the walking speed. However, the gait length change is not possible while tracking fixed predefined joint trajectories. Hence, in this paper the NMPC is designed in such a way that it has the ability to change the gait length appropriately. In addition, some schemes will be proposed to reduce the computation time of the NMPC. Simulating results show good performance of the proposed method in trajectory-free walking of biped robots as well as disturbance rejection.