摘要
针对喷涂机器人离线轨迹规划系统中路径顺序与喷涂方向同时影响喷涂效率的特点,将喷涂路径的组合与排序问题建模成开环式广义旅行商问题,并建立了相应的代价矩阵与优化目标;提出了一种基于分布式估计的路径组合优化算法,该算法在遗传算法中引入统计学习的手段,采用基于概率的模型学习和采样算法实现更好的进化效率,从而能够更加有效地获得全局最优解。通过多组数据的仿真,验证了该算法解决路径组合问题的有效性与可行性。
Considered the characteristics of the path sorting and integration problem in off-line trajectory planning system of spray painting robots,it modeled the integration problem as an open generalized traveling salesman problem （GTSP）, and established the corresponding cost matrix and optimization objective. In order to solve the problem, this paper proposed an optimization algorithm based on distributed estimation. The algorithm introduced statistical learning into genetic algorithm, and used learning based on the probability model and sampling algorithm to replace the crossover and mutation in genetic algorithm to achieve the evolution of population. So it can get the global optimum effectively. Results of simulations verify the effectiveness and feasibility of this method.
出处
《计算机应用研究》
CSCD
北大核心
2012年第8期2935-2938,共4页
Application Research of Computers
基金
国家重大科技专项基金资助项目(2010ZX04008-041)
关键词
喷涂机器人
路径组合
广义旅行商问题
分布估计
painting robot
path sorting
generalized traveling salesman problem（GTSP）
distribution estimation
