Abstract: A Lagrangian Support Vector Machine （LSVM） based robust adaptive controller is proposed for the trajectory tracking of robot manipulators. With gradient projection learning method, LSVM is used to compensate for the unknown part of the robot manipulator system. An adaptive law based on the Lyapunov stability theory is designed to further adjust the weights online. Moreover, a robust term is added to guarantee the stability and robustness of the controlled system. The effectiveness of the proposed approach is illustrated by the simulation results on a two-link manipulator.

Key words: Lagrangian support vector machines, robot manipulators, robust control, gradient projection method

摘要： 提出了一种基于Lagrangian支持向量机的不确定机械手鲁棒自适应控制方法。Lagrangian支持向量机采用梯度投影法学习机械手系统的未知部分，来对机械手系统进行非线性补偿。根据Lyapunov稳定性理论设计自适应律进一步在线调整支持向量机的参数，并叠加一个滑模控制项，以保证控制系统的稳定性和鲁棒性。对两关节机械手的仿真结果证明了以上控制方法的有效性。

关键词: Lagrangian支持向量机, 机械手, 鲁棒控制, 梯度投影法