摘要
针对不确定、时变和非线性机器人系统的实时性要求,提出了采用滑模变结构和RBF神经网络相结合来构造控制器。用带有符号函数的滑模变结构控制器来产生一个控制输入信号,同时利用具有快速学习能力的RBF神经网络来学习外界的不确定性,增强系统的自适应能力,使之达到更佳的控制效果,并在文中证明了系统的稳定性。最后给出了对两连杆机器人的仿真,验证了控制效果。
A novel robust adaptive control strategy based on sliding variable structure and RBF neural network is presented under considering the uncertainties, real-time and nonlinear performances of robot system. This paper uses the sliding variable structure with signum function to generate control input signals and RBF neural network as an auxiliary controller in order to eliminate the effects of uncertainty system error. This paper also proves the stability of the control system. Finally, simulations are given for a two-link robot, and validates the control performance.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2006年第z3期1881-1882,共2页
Chinese Journal of Scientific Instrument
基金
河北省科技攻关项目(04547007D)
