Abstract

In order to improve the control accuracy between the joints of the manipulator for non-linear uncertain systems such as single-joint manipulators. The \({{{\text{D}}}^{\alpha }}\) type and \({\text{P}}{{{\text{D}}}^{\alpha }}\) type fractional order iterative learning control (ILC) strategies are proposed based on combining fractional order calculus and ILC. The experimental verification was carried out with the two-joint manipulator as the research object. The experimental results are as follows: the fractional order ILC is compared with the traditional ILC, in which the minimum value of the position error of the \({{{\text{D}}}^{\alpha }}\) type fractional order ILC is reduced by 0.00016 and 0.00111 rad compared with the traditional \({\text{D}}\) type ILC, and the \({\text{P}}{{{\text{D}}}^{\alpha }}\) type fractional order ILC compared with the traditional \({\text{PD}}\) type ILC, the minimum value of the position error is reduced by 0.00022 and 0.00024 rad respectively. It shows that the algorithm proposed in this paper can improve the tracking performance of the system and improve system control accuracy.

中文翻译：

---

基于分数阶迭代学习的机械手控制策略

摘要

为提高单关节机械手等非线性不确定系统的机械手关节间控制精度。的\（{{{\文本{d}}} ^ {\阿尔法}} \）类型和\（{\文本{P}} {{{\文本{d}}} ^ {\阿尔法}} \）基于分数阶微积分和ILC的类型，提出了一种分数阶迭代学习控制（ILC）策略。以双关节机械手为研究对象进行了实验验证。实验结果如下：分数阶ILC与传统ILC比较，其中\({{{\text{D}}}^{\alpha }}\)类型的位置误差的最小值分数阶 ILC 比传统的减少了 0.00016 和 0.00111 rad\({\text{D}}\)型 ILC，而\({\text{P}}{{{\text{D}}}^{\alpha }}\)型分数阶 ILC 与传统的\({\text{PD}}\)型ILC，位置误差的最小值分别降低了0.00022和0.00024 rad。表明本文提出的算法能够提高系统的跟踪性能，提高系统控制精度。