基于RRTConnect算法的双履带起重机路径规划研究

发布时间：2018-06-06 15:09

本文选题：路径规划 + 双机协同吊装　；参考：《大连理工大学》2012年硕士论文

【摘要】：被吊设备、起重机以及吊装方式的多样化、复杂化等特点决定了起重机吊装是一类高危险性的活动。与单机吊装作业相比,两台履带起重机(文中简称双机)协同吊装的危险系数更高。为了安全顺利地完成吊装作业,合理高效的起重机路径规划就显得尤为重要。为此,本课题将针对双机协同吊装任务,抽象问题的数学模型,对闭环约束、非完整约束等开展深入研究,应用并改进RRTConnect算法,提出可以应用于平移和翻转被吊设备的通用方法,规划出满足无碰撞、实时保证绳偏角、履带起重机行走特性和起重性能等要求的优化路径。首先,通过对吊装存在的安全性问题进行分析,以及对起重机吊装路径规划研究现状进行总结,论述了本课题的研究意义,同时对路径规划方法的特点进行研究确定应用和改进RRTConnect(双向多步扩展快速随机搜索树)算法来解决双机路径规划问题。其次,在讨论机器人路径规划的基础上,对双机路径规划问题进行深入剖析,构建双机吊装系统数学模型,并对数学模型的变量进行详细说明。再次,对双机吊装中两类常见吊装任务即将被吊设备翻转和将被吊设备平移,论述了采用RRTConnect算法解决双机路径规划问题的步骤,同时给出了逆向运动学确定双机初始和目标位姿的计算过程和对路径的优化处理方法,并用三个典型案例验证方法的可行性。最后,对RRTConnect算法做改进,提出了基于关键位姿的RRTMultiPhrase算法,以解决双机吊装中较为复杂的吊装任务路径规划,论述了RRTMultiPhrase算法的原理,同时给出了关键点选择的一般方法,接着对两个复杂案例分别用三种方法进行路径规划,说明本文提出算法的可行性。本文分析了双机吊装作业的特点,对给定双机初始和目标位姿的协同吊装路径规划问题展开深入的探索和研究,实现了考虑直线行走、转弯、起升、回转和变幅等动作并满足无碰撞、闭环约束、非完整约束、不超载等约束的路径规划,解决了前人规划方法不适用于需要起重机行走或者使被吊设备翻转的吊装任务的局限性,并且将提出的方法应用于典型的吊装案例,证明其可行性、实用性和高效性。
[Abstract]:The diversity and complexity of hoisting equipment, cranes and hoisting methods determine that crane hoisting is a kind of high dangerous activity. Compared with single crane hoisting, the hazard coefficient of two crawler cranes is higher than that of single crane. In order to complete the hoisting operation safely and smoothly, it is very important to plan the crane route reasonably and efficiently. For this reason, this subject will carry on the thorough research to the double machine cooperative hoisting task, the abstract mathematical model, the closed loop constraint, the nonholonomic constraint and so on, and applies and improves the RRT Connect algorithm. A general method which can be applied to the translation and reversal of hoisting equipment is proposed to plan the optimal path to meet the requirements of non-collision, real-time guarantee of rope deflection angle, crawler crane travel characteristics and hoisting performance, and so on. By analyzing the safety problems existing in hoisting and summarizing the research status of crane hoisting path planning, the research significance of this subject is discussed. At the same time, the characteristics of the path planning method are studied and applied, and the RRTConnect algorithm is improved to solve the dual machine path planning problem. Secondly, based on the discussion of the robot path planning, the RRTConnect algorithm is improved to solve the dual-machine path planning problem. In this paper, the path planning problem of two machines is analyzed in depth, and the mathematical model of the hoisting system is constructed, and the variables of the mathematical model are explained in detail. For the two kinds of common hoisting tasks which are about to be overturned by the hoisting equipment and the translation of the hoisting equipment, the steps of using RRT Connect algorithm to solve the dual machine path planning problem are discussed in this paper. At the same time, the calculation process of inverse kinematics to determine the initial and target orientation of two machines and the method of path optimization are given, and the feasibility of the method is verified by three typical cases. Finally, the RRTConnect algorithm is improved. The RRT MultiPhrase algorithm based on the key position and pose is proposed to solve the more complicated path planning of hoisting task in the dual computer hoisting. The principle of the RRT MultiPhrase algorithm is discussed, and the general method of selecting the key points is given. Then, two complex cases are planned with three methods, which shows the feasibility of the algorithm in this paper. This paper analyzes the characteristics of double-machine hoisting operation. In this paper, the cooperative hoisting path planning problem for the given initial and target positions of two machines is deeply explored and studied, and the actions such as straight line walking, turning, hoisting, turning and amplitude shifting are considered, and the collision free, closed loop and nonholonomic constraints are satisfied. The path planning with no overload and other constraints solves the limitations of the previous planning methods which are not suitable for hoisting tasks that require cranes to walk or make the lifting equipment flip over, and the proposed method is applied to typical hoisting cases to prove its feasibility. Practicality and efficiency.
【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

【相似文献】