机器人非线性连续预测控制

研究了一种机器人非线性连续预测控制方案，其预测模型通过对系统状态进行泰勒级数展开并作适当的截改进来获得，并通过优化性能指标求得控制律。与其他非线性模型预测控制相比，所得到的器具有解析形式，方法简单、计算量小、利用在线应用。仿真结果表明了非线性预测控制器的有效性。

Nonlinear continuous predictive controller for robot manipulator

A time continuous nonlinear model predictive control scheme for robot manipulator is studied with a predictive model acquired by truncating appropriately Taylor series expansion for system states, and control law identified by optimizing the cost function. Compared with other nonlinear predictive control using such empirical predictive model as Hammerstein model, Volterra model, NARMAX model and so on, the resulting controller, which has analytical formula with less calculation, is simple and fit for on line application. Simulation results show the nonlinear predictive controller is very effective.