基于滑模补偿的六轴机械臂RBF网络自适应控制

提出了一种基于滑模补偿的RBF网络自适应控制方法,并将其用于六轴机械臂上,实现了在模型不确定情况下高精度的位置跟踪以及快速的逼近速度.为了更好地保证其在摩擦力、外部干扰误差和参数变化等因素影响下的稳定性,采用滑模控制作为辅助控制,对系统进行鲁棒补偿,并且用模糊控制对切换增益进行时变以更好地补偿不确定项.所设计的自适应律权值不断进行在线调整,并应用Lyapunov定理验证了其稳定性.通过仿真结果和与其他文献方法进行的比较证明了所给出方法收敛速度更快、鲁棒性更强.同时也证明了在实际工程中,实际建模参数与理想值相差很大.

RBF Network Adaptive Control Based on SMC Compensation for Six-axis Manipulator

A RBF neural network adaptive control based on sliding mode control compensation is proposed， and applied to the six-axis robot manipulator， to achieve high-precision position tracking and rapid approaching speed in the case of model uncertainty. Sliding mode control is adopted as auxiliary for system robust compensation and fuzzy control is used for real-time switching gain changing to make the uncertainties better， in order to keep the system′s stability toward the influence of friction， external interference errors and parameter changes. Adaptive laws weights were adjusted constantly online， and Lyapunov theorem was used to prove the stability. The simulation result indicated that a faster convergence rate and stronger robustness can be acquired with the proposed control algorithm， comparing with other research. It also shows that in reality， significant difference exists between the modeling parameters and the actual values.