蜂群粒子群混合算法在机械手参数优化中的应用研究  

Study on Application of Hybrid Optimization Algorithm in Parameters Optimization of Manipulator

作　　者：宫赤坤[1] 孙英慧[1] 袁立鹏 

机构地区：[1]上海理工大学机械工程学院,上海200093 [2]常州恐龙园股份有限公司,常州213022 [3]哈尔滨工业大学机电工程学院,哈尔滨150001

出　　处：《系统仿真学报》2018年第1期105-113,共9页Journal of System Simulation

基　　金：国家科技支撑计划(2013BAH57F00)

摘　　要：工业机械手进行精密加工时,尽可能在全姿态工作空间内部完成。为了得到紧凑的结构,同时满足加工点都为全姿态点,提出了一种新的蜂群粒子群混合优化算法。建立机械手全姿态空间的结构模型,引入粒子群中全局最优粒子和惯性权重两个参数,通过中间值进行分类,差异权衡惯性权重,增强寻优能力。经验证,该算法搜索能力更强,寻优精度更高。对结构优化模型进行参数辨识,结果表明,在保持原来全姿态空间大小基本不变的情况下得到了结构紧凑、质量最优的杆长参数,从而验证了所提算法的有效性、可行性。When the industrial robot carries on the precise processing, it is possible to finish in the work space of the whole attitude. In order to obtain the compact structure and meet the processing point, a new hybrid optimization algorithm is proposed. The structure model of the robot is established, and the global optimal particle and inertia weight of the particle swarm is introduced. The two parameters are classified by the intermediate value. The dynamic changes of the search medium and the optimization ability are enhanced. After verification, the algorithm has better search ability and higher accuracy. Finally, parameter identification of the structural optimization model, the results show that the obtained structure is compact and the quality of the rod length parameter is better while keeping the same size of the original quasi total attitude, Thereby, the proposed algorithm is effective and practical.

关 键 词：工业机械手 全姿态工作空间 杆长参数 中间值 惯性权重 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]