基于改进遗传算法的虚拟装配路径规划研究

Study of Virtual Assembly Path Planning Based on Improved Genetic Algorithm

装配路径规划是虚拟装配技术的核心,为了获得最优装配路径,在凸四边形障碍物环境中,通过对初始种群选取方法的重新定义,保证了参加演化算法的每条染色体都是可行的,同时对传统遗传算法中的变异算子和杂交算子进行改进,提高了算法的有效性。最后实验证明将改进的遗传算法运用到机械式减速箱装配中可得到更优的装配路径。

Assembly path planning is the core of virtual assembly. In order to obtain the optimal path,an improved algorithm based on genetic algorithm is proposed in the convex quadrilateral obstacle environment. The selection of the sample population is redefined to make sure that all the chromosomes involved in the evolutionary algorithms are useful. And the traditional mutation operator and crossover operator of genetic algorithm are improved to enhance the efficiency of the algorithm. Experiment shows that the mechanical reducer can be used to optimize the assembly path.