With increasing demand for using robotic manipulators in industrial applications, controllers specific for performing repeatable tasks are required. These controllers must also be robust to model uncertainties. To address this research issue, a repetitive learning control method fused with adaptive fuzzy logic techniques is designed. Specifically, modeling uncertainties are first modeled with a fuzzy logic network and an adaptive fuzzy logic strategy with online tuning is designed. The stability is investigated via Lyapunov type techniques where global uniform ultimate boundedness of closed loop system is guaranteed. Numerical simulation results obtained from a two degree of freedom robot manipulator model and experiments performed on a robot manipulator demonstrate the efficacy of the proposed control methodology.

随着在工业应用中使用机器人操纵器的需求不断增加，需要专门用于执行可重复任务的控制器。这些控制器还必须对不确定性建模具有鲁棒性。为了解决这一研究问题，设计了一种与自适应模糊逻辑技术融合的重复学习控制方法。具体而言，首先使用模糊逻辑网络对建模不确定性进行建模，然后设计具有在线调整的自适应模糊逻辑策略。通过Lyapunov型技术研究了稳定性，其中保证了闭环系统的整体一致最终有界性。从两自由度机器人操纵器模型获得的数值模拟结果以及在机器人操纵器上执行的实验证明了所提出的控制方法的有效性。