论文部分内容阅读
针对复杂环境中的大型辅助工装,本文在非均匀环境建模的基础上,提出了一种搜索工装可行路径的改进型蚁群算法.首先,在环境建模过程中,利用凸投影计算环境中的障碍物在投影平面的障碍区域,然后利用Minkowski和对其进行拓展,并利用线性四叉树法实现环境的非均匀栅格建模.其次,在进行路径规划时,提出一种改进型蚁群算法对工装的可行路径进行搜索.该算法采取了以下策略以提高性能:融合了最大最小蚁群(MMAS)算法与蚁群系统(ACS)算法的信息素更新方式,在栅格选择策略中引入了目标距离和障碍物距离等启发式信息,设计了与已建立的环境模型相符合的适应度函数.最后,以大型激光驱动器的靶场环境为对象,对本文算法的有效性进行了验证,证明本文算法能够在存在多障碍物的复杂环境中高效完成工装运动路径的规划.设计了对比实验,将本文算法与基本蚁群算法进行了对比,实验结果表明,与基本蚁群算法相比,本文算法具有更快的收敛速度,规划结果具有更高的适应性.
In this paper, based on the modeling of non-uniform environment, an improved ant colony algorithm is proposed to search for the feasible path of tooling.Firstly, in the process of environment modeling, using the convex projection computing environment Of the obstacle in the projection plane of the obstacle area, and then use Minkowski and its expansion, and the use of linear quadtree method to achieve environmental non-uniform grid modeling.Secondly, in the path planning, an improved ants Group algorithm to search the feasible path of tooling.The algorithm adopts the following strategies to improve the performance: the pheromone updating method which combines MMAS algorithm and ACS algorithm is applied in the grid selection strategy The heuristic information, such as target distance and obstacle distance, is introduced and the fitness function which is consistent with the established environmental model is designed.Finally, the effectiveness of the proposed algorithm is validated by the shooting range of large laser driver, It proves that the algorithm of this paper can efficiently plan the path of tooling movement in complicated environment with many obstacles.A comparison experiment is designed to compare the algorithm with the basic ant colony The experimental results show that compared with the basic ant colony algorithm, the proposed algorithm has a faster convergence rate and the planning result has a higher adaptability.