摘要
首先用滑模控制策略对被控对象进行控制仿真,在分析结果后,结合神经网络滑模控制方法,充分利用神经网络滑模控制的学习能力强,自适应辨识能力强,可以无穷逼近任意函数的优点。仿真结果表明,神经网络滑模控制方法的跟踪效果好,系统误差小,可以满足机器人控制的要求,能够解决机器人的轨迹跟踪问题,仿真实验证实了该控制策略的正确性和有效性。
Firstly,by the use of the sliding control strategy to simulate the object, after analysing the results, combining the neural network sliding control method, making full use of the learning ability of the neural network sliding control and the ability of adaptive identification ,and it also can approximate any function infinite advantages. The simulation results showed that the neural network control method of tracking sliding took good effect, the system error was small, satisfing the requirements of robot control, solving the problem of robot tracking. The simulation experiments confirmed the correctness of the control strategy and effectiveness.
出处
《微特电机》
北大核心
2008年第8期32-35,共4页
Small & Special Electrical Machines
基金
国家自然科学基金资助项目(50375067)
江苏省教育厅资助项目(03KJD510072)
关键词
并联机器人
滑模控制
神经网络
仿真
控制策略
parallel robot
sliding control
neural network
simulation
control strategy
