遗传禁忌搜索算法在工业机器人结构参数辨识上的应用  

Application of Genetic-Tabu Search Algorithm in Industrial Robot Structural Parameter Identification

作　　者：熊杰[1,2] 杨东升[2] 王允森[1,2] 

机构地区：[1]中国科学院大学,北京100049 [2]中国科学院沈阳计算技术研究所,沈阳110168

出　　处：《组合机床与自动化加工技术》2015年第12期4-7,11,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：核高基专项(2012ZX01029-001-002)

摘　　要：针对工业机器人标定过程中的结构参数辨识方程形式复杂、计算繁琐等问题,提出一种采用遗传禁忌搜索算法来寻求辨识方程的最优解的方法。首先采用遗传算法做全局搜索,为加快收敛速度,遗传算法的选择算子、交叉算子和变异算子的参数按照动态适应的方式进行。遗传算法调用结束后,将种群的最优个体作为禁忌搜索的初始解,调用禁忌搜索算法做进一步局部搜索。为了提高解的精确性,邻域半径根据适应度值进行动态调整。仿真实验结果表明遗传禁忌搜索算法能够有效的得到结构参数的完全辨识。For structural parameter identification problems w ith industrial robot calibration,considering the identification equations tends to form complexes,tedious calculation,so a genetic-tabu search algorithm（ GTSA） for the optimal solution of identification equation w as proposed. Firstly,genetic algorithm（ GA） w as used to global search. To accelerate the convergence speed of GA,the selection operator,crossover operator,mutation operator parameters w as deal w ith the dynamic adaptive w ay. After GA called,the best individual in the population w ould be treated as the initial solution to tabu-search（ TS）. Then TS w as called for further local search. In order to improve the accuracy of the solution,the neighborhood radius dynamically adjusted based on the fitness value. Through simulation experiment,the results show that GTSA can effectively get the full identification of structural parameters.

关 键 词：工业机器人 参数辨识 遗传算法 禁忌搜索 

分 类 号：TH166[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]