基于模拟退火算法的Stewart平台的运动学正解分析

2020-08-13 07:17:51邵云龙吴建民孔森
软件 2020年6期

邵云龙 吴建民 孔森

摘  要: 在工业生产之中Stewart平台被广泛应用。由于上平台得位姿不易得到测量,又需要监控机构的运行位置。因此对Stewart平台进行运动学正解分析是十分必要的。本文利用牛顿-泰勒法建立Stewart平台的运动学正解模型,求解出位置逆解、速度逆解。并利用模拟退火算法与初值无关并且具有概率突跳性和Newton-Raphson迭代法精度高的特点,提出了模拟退火算法和Newton-Raphson迭代法相结合的Stewart平台位置正解方法。通过MATLAB编制了运动学正解和运动学反解的程序。利用SolidWorks进行建模,并且在Adams上进行运动仿真验真了算法的准确性, 为下一步的研究提供理论依据。

关键词: Stewart平台;模拟退火算法;Newton-Raphson;SolidWorks;Adams

中图分类号: TP273    文献标识码: A    DOI:10.3969/j.issn.1003-6970.2020.06.051

本文著录格式:邵云龙,吴建民,孔森,等. 基于模拟退火算法的Stewart平台的运动学正解分析[J]. 软件,2020,41(06):247252+256

【Abstract】: The Stewart platform is widely used in industrial production due to its advantages such as high rigidity, stable structure, and other advantages. However, since the posture of the upper platform is difficult to measure during the operation, it is necessary to monitor the operating position of the mechanism to ensure the safe operation of the mechanism. Therefore, it is necessary to analyze the forward kinematics of the Stewart platform. In this paper, the Newton-Taylor method is used to establish a positive kinematics solution model of the Stewart platform, and the inverse position and velocity solutions are solved. The simulated annealing algorithm is independent of the initial value and has the characteristics of probability jump and high accuracy of the Newton-Raphson iteration method. A Stewart platform forward solution method combining the simulated annealing algorithm and the Newton-Raphson iteration method is proposed. Programs for positive and negative kinematics were compiled by MATLAB. Modeling with SolidWorks and motion simulation on Adams verified the accuracy of the algorithm, providing a theoretical basis for further research.

【Key words】: Stewart platform; Simulated annealing algorithm; Newton-Raphson; SolidWorks; Adams

0  引言

Stewart平台是自动化领域最具标志性的并联机械装置,在工业上得到了广泛应用,如舰载稳定平台、海浪运动模拟器、人形机器人腰部结构等等。近年来,由于运动学正解问题在研究过程中处于基础性地位,越来越多的国内外学者投身于对其的研究[1]。对于Stewart平台,上平台可以实现6个自由度的运动。为了使上平台可以更加精确的完成预先设定的运动轨迹,其运动学正解研究变得愈发重要起来。Manuel Cardona将牛顿-拉夫森数值迭代算法应用于Stewart平台运动学位置正解的过程中,能够快速得到运动学正解的其中一解[2]。张宗之,秦俊奇,陈海龙,刘平松等人采用BP神经算法,解决了数值解法当中出现的计算速度较慢,迭代初值选取不当将会导致运算结果出现较大偏差的不利情况[3]。……

登录APP查看全文