Decentralized Robust Tracking Control for Robots with Uncertain Parameters

石宗英,钟宜生,徐文立
DOI: https://doi.org/10.3321/j.issn:1001-0920.2004.07.009
2004-01-01
Abstract:A new design method of decentralized robust tracking controller for robots with uncertain parameters is proposed. The dynamics of the joint subsystems are divided into two parts: an artificial nominal model and a nonlinear time-varying uncertain model. Then a corresponding nominal controller and a robust compensator are designed. The nominal controller enables the nominal closed-loop system to achieve desired tracking properties, and the robust compensator reduces the influence of the parameter uncertainties and the nonlinear couplings among the joints. Only using local position feedbacks, the controller has the advantage of easy realization and tuning on-line. The simulation results show the effectiveness of the method.
What problem does this paper attempt to address?