The dominant physical phenomena in hydraulic drives are generally well known, why the model equations describing the dominant dynamics may be established with a high level of certainty. To some extend, this is also the case for the model parameters when these are based on data sheet information. However, parameters such as the effective bulk modulus, leakage, external disturbances etc. may be difficult to evaluate, and may furthermore be varying. In regard to control design, linear methods may be difficult to apply and stability margins difficult to evaluate, unless dynamic models are established prior to the control design. This problem may be overcome using adaptive controllers, adjusting themselves to uncertainties/variations. This often shifts the problem from adjusting the controller parameters, to adjusting the parameters of the control parameter adaption mechanism, which in many cases involves a significant number of parameters. This paper considers a novel adaptive control algorithm, theoretically applicable to systems of arbitrary orders, and potentially only with three tuning parameters. The proposed algorithm is considered in relation to the position control of a hydraulic winch drive, and results imply that excellent motion tracking performance may be achieved utilizing only state feedback.
Online Sparse Gaussian Process (GP) Regression Model For Human Motion Tracking
