东篱科研大数据发现系统（DRDS）

位置：成果数据库 > 期刊 > 期刊详情页

Self-adapting control parameters modified differential evolution for trajectory planning of manipulators

ISSN号：2095-6983
期刊名称：《控制理论与技术：英文版》
时间：0
分类：TP13[自动化与计算机技术—控制科学与工程;自动化与计算机技术—控制理论与控制工程]
作者机构：[1]College of Information and Electric engineering, Hunan University of Science and Technology, Xiangtan Hunan 411201, China, [2]College of Electric and Information Engineering, Hunan University, Changsha Hunan 410082, China
相关基金：This work was supported by the National Natural Science Foundation of China （No.60375001）, and the High School Doctoral Foundation of China （NO. 20030532004）.

作者： Lianghong WU[1,2], Yaonan WANG[2], Shaowu ZHOU[1]

关键词：控制参数, 微分演化, 控制机械手, 轨道, Self-adapting control parameters, Differential evolution, Redundant manipulator, Trajectory planning

中文摘要：

原来的微分进化(DE ) 的控制参数被使全部进化进程全部固定了。然而，它不是到在为不同优化问题的 DE 的适当地设置的控制参数的一项容易的任务。根据选择在寻找的地方产生差别向量的二不同单个向量的相对位置，为差别向量的规模因素 F 的自我改编的策略被建议。以在当前的人口的目标向量的集中地位，自我改编的转线路概率常数 CR 策略被建议。因此，当更坏的目标向量有大 CR 时，好目标向量有更低的 CR。同时，变化操作员被修改改进集中速度。这些建议途径的表演与一些基准问题的使用被学习；适用于一个三关节的冗余的操纵者计划的轨道。最后，实验结果证明建议途径能极大地改进坚韧性；集中速度。

英文摘要：

Control parameters of original differential evolution （DE） are kept fixed throughout the entire evolutionary process. However, it is not an easy task to properly set control parameters in DE for different optiinization problems. According to the relative position of two different individual vectors selected to generate a difference vector in the searching place, a self-adapting strategy for the scale factor F of the difference vector is proposed. In terms of the convergence status of the target vector in the current population, a self-adapting crossover probability constant CR strategy is proposed. Therefore, good target vectors have a lower CFI while worse target vectors have a large CFI. At the same time, the mutation operator is modified to improve the convergence speed. The performance of these proposed approaches are studied with the use of some benchmark problems and applied to the trajectory planning of a three-joint redundant manipulator. Finally, the experiment results show that the proposed approaches can greatly improve robustness and convergence speed.

同期刊论文项目