【作者】 陈亮；

【导师】 宋玉阶；

【作者基本信息】 武汉科技大学 ， 控制理论与控制工程， 2010， 硕士

【摘要】 本文讨论了目前在焊接机器人路径规划领域广泛使用的智能算法:遗传算法和Hopfield神经网络。针对这些算法处理路径优化问题时的不足,提出了基于DNA计算处理焊接机器人路径规划问题的算法。基于焊接机器人路径规划问题的数学模型,将焊接机器人路径规划问题转化为完全正权无向图的最短Hamilton环路问题;结合焊接机器人工作的实际情况,运用DNA算法获得最短Hamilton环路路程,从而实现焊接机器人的工具原点运动的最优化;最后通过算例证明了方法的有效性。针对已有的DNA计算模型的缺点,采取了相应的改进,把包含重复顶点的闭链剔除出去,满足了实际规划问题中不重复焊接同一焊点的要求。本文还讨论了DNA计算实现实际工作中焊接机器人路径规划时出现的几个问题,并提出了作者的一些想法,为以后研究这些问题指出了努力的方向。更多还原

【Abstract】 Discussion of the current field of welding path widespread use of Intelligent algorithms: genetic algorithms and Hopfield neural networks. These algorithms address the lack of path optimization problems is proposed welding robot path planning algorithms based on DNA computing.Based on the mathematical model, the welding robot path planning problem is transformed into the shortest completely weighted undirected graph of Hamilton loop problem; combination of the actual situation of the robot, the use of DNA method was the shortest Hamilton Circle distance to the origin of welding robot movement optimization tool; Finally, a numerical example demonstrates the effectiveness of the method. DNA computing model for the existing shortcomings, take a corresponding improvement, to include removing duplicate vertices of closed-chain out to meet the actual planning problems do not repeat the same solder welding requirements.This article also discusses the practical work of DNA Computing for welding robot path planning, several issues emerged and made of some of the ideas for the future study of these issues that direction.更多还原