履带式起重机分段路径规划算法的设计与实现

发布时间：2018-05-10 06:13

本文选题：移动式起重机 + 路径规划　；参考：《大连理工大学》2012年硕士论文

【摘要】：重要设备现场吊装方案设计和吊装路径规划对于保障安全和提高效率具有十分重要的意义。吊装方案设计过程中,工程师需要考虑起重机在施工场地中的运行轨迹,如果工作环境比较复杂,起重机通过能力较低,手工计算很难找到一条无碰撞且合理的路径。履带式起重机路径规划问题属于高维多约束路径规划问题,当前还没有一种有效的算法来解决该问题,工程实践中仍然主要依靠工程师的经验和手工计算来确定方案中起重机的路径。随着虚拟仿真技术的发展,吊装行业也开始借助仿真软件来加速其吊装方案的设计,出现了很多基于虚拟仿真环境的规划算法,但是目前提出的起重机运动规划算法大多未考虑行走,因为考虑起重机的行走,不仅在C空间中增加了3个维度,而且行走会引入非完整性约束,会大大增加路径规划的难度。本文在总结已有算法的基础上,根据履带起重机的运动特点作出相应的分析和改进,设计并实现了单台起重机的分段路径规划算法,将下车行走路径与上车路径进行解耦,其中,下车规划阶段只处理起重机的下车运动,采用导航网格作为底层的搜索空间描述,并使用A*算法进行寻路、funnel算法进行修正：上车规划阶段规划起重机的上车动作,采用RRT算法并进行了适当的变种以适应局部路径规划。新提出的算法适应于单台履带式起重机的路径规划,碰撞检测次数少,规划成功率高,速度快。本文基于该算法开发了起重机路径规划系统,并将其作为子系统集成进一款吊装仿真软件中,大大提升了该款软件的市场竞争力。
[Abstract]:It is very important to design the hoisting scheme and plan the hoisting path for the important equipment to ensure the safety and improve the efficiency. In the design of hoisting scheme, the engineer needs to consider the running track of crane in the construction site. If the working environment is more complicated and the crane's passing capacity is low, it is difficult to find a collision-free and reasonable path by manual calculation. The path planning problem of crawler crane belongs to the high-dimensional multi-constraint path planning problem, but there is no effective algorithm to solve the problem. Engineering practice still relies on engineer's experience and manual calculation to determine the route of crane in the project. With the development of virtual simulation technology, the hoisting industry began to use simulation software to accelerate the design of its hoisting scheme, and many planning algorithms based on virtual simulation environment appeared. However, most of the proposed crane motion planning algorithms do not consider walking, because considering the traveling of crane, not only three dimensions are added in C space, but also non-integrity constraints are introduced into walking, which will greatly increase the difficulty of path planning. On the basis of summarizing the existing algorithms, according to the movement characteristics of crawler crane, this paper designs and implements a piecewise path planning algorithm for a single crane. In the stage of getting off the train, only the movement of the crane is dealt with, the navigation grid is used as the bottom search space description, and the algorithm A * is used to correct the route finding algorithm: in the planning stage of getting on the train, the crane is planned to get on the train. RRT algorithm is used to adapt to local path planning. The new algorithm is suitable for the path planning of single crawler crane, with less collision detection times, higher planning success rate and faster speed. This paper develops a crane path planning system based on this algorithm, and integrates it as a subsystem into a hoisting simulation software, which greatly improves the market competitiveness of the software.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH213.7

【参考文献】