| Robot is a high complicated time-varied, strong-coupled, nonlinear system, actually, it has various kinds of uncertainties, such as measurement errors, frictions, varying load, random disturbances, unmodeled dynamics and so on. So it's hard to get the orbicular, precious model of robot system. Aims at these situations above, the paper imports the dissipative theory and puts forward the robust H_∞control method which based on Backstepping design art. Besides, we also give another control method which called MBDA method.H_∞control is based on the principle of restraining disturbance. Insuring the systems'stable, H_∞control makes systems disturbance have a demanded lowest influence to outputs. Based on H_∞theory and considering the relationship of parameters uncertainty and the outer disturbance, this paper brings forward two control methods, one is that control the parameters uncertainty and the outer disturbance respectively; the other is regards the parameters uncertainty as a part of the outer disturbance, by combining the Backstepping method and H_∞control theory, it avoids the calculation of HJI equation in controlling nonlinear systems using H_∞method, simplifies the calculation course of the controller and the required controller has a simply structure.MBDA ( disturbance restrain which based on models ) control method which is very simple and practical is first put forward by KOREA scholar Choi several years ago. It's main idea is that to modeling a parallel model for practical model, then, combine the practical model and parallel model in one in a ordered way by using PD and PID controllers. In this paper, the MBDA theory is applied into robot control problem, not only the method resolves the error problems caused by unknowing the precision model, but also it's simple arithmetic and easy to carry out . More important, the method has the excellent ability of disturbance attenuation and robustness. |
