混凝土喷射机多关节臂架运动学建模与控制算法的研究

发布时间：2018-02-11 11:17

本文关键词： 喷射机冗余自由度梯度投影多目标优化闭环控制　出处：《哈尔滨工业大学》2013年硕士论文　论文类型：学位论文

【摘要】：混凝土喷射机是一种通过喷射速凝混凝土进行开挖断面支护的工程机械，目前广泛应用于隧道、公路和煤矿等施工工程中。相比于传统的人工混凝土喷射施工方法，混凝土喷射机降低了工人的劳动强度、提高了施工效率和施工质量。但是，现阶段使用的大多数喷射机只能对臂架进行单关节的控制，操作复杂性较高，难以保证较高的喷射质量。随着工业自动化程度的不断提高，不论是在学术上还是实际的工程中，具有智能控制功能，能实现多关节联动控制的喷射机臂架，正成为研究热点。本文将喷射机臂架的控制机器人化，对臂架运动学展开研究，以注重工程的开发应用及其理论研究的探索作为研究目的，拟解决以下几个方面的问题：首先，建立现有喷射机臂架的运动学模型。喷射机臂架是一个液压驱动机械臂，不同的关节其驱动方式不一样，其运动学关系也不一样。通过对臂架结构的详细分析，对其做相应的简化，得到臂架的运动学模型，并最终建立了关节的驱动空间、关节空间和臂架末端任务空间三者之间的运动学关系。为进行臂架系统的驱动空间到末端任务空间的运动设计与规划提供了基础。其次，，在不考虑末端喷射头的情况下，喷射机臂架一共包括七个自由度，含有一个冗余自由度，其运动学逆解非常复杂。本文结合喷射机臂架结构和实际施工情况，设计了能同时对关节运动位置和速度进行优化的运动学逆解算法。该算法主要基于梯度投影法，通过对各个优化目标的加权处理，解决了多目标优化问题，实现了在保证臂架运动平稳的前提下，防止关节位移的超限的控制目标。最后，由于梯度投影算法是一种基于雅克比矩阵伪逆的控制算法，因此，在计算机控制系统进行算法实施的过程中，不可避免的产生迭代误差。本文对这种算法的不精确性做了详细的研究，并提出了一种闭环控制策略，使误差收敛。综上所述，本文建立了喷射机臂架的运动学模型，对运动学问题进行了详细的探讨，设计了一种逆运动学算法。并且，通过对控制算法的改进，提出一种闭环控制策略，保证了臂架末端跟踪的精确性。算法整体的设计方案和研究结果可为同类机械臂架提供研究的理论依据和实践参考。
[Abstract]:The concrete ejector is a kind of construction machinery which is used to support the excavation section by shotcrete, which is widely used in tunnel, highway, coal mine and other construction projects. Compared with the traditional artificial concrete shotcrete construction method, concrete ejector is widely used in the construction of tunnel, highway and coal mine. The concrete ejector reduces the labor intensity of the workers and improves the construction efficiency and construction quality. However, most of the ejectors used at the present stage can only control the boom with a single joint, and the operation complexity is high. It is difficult to guarantee high injection quality. With the continuous improvement of industrial automation, no matter in academic or practical engineering, it has intelligent control function and can realize multi-joint linkage control of ejector boom. In this paper, the control robot of the boom of ejector and the kinematics of the boom are studied. The purpose of this research is to pay attention to the development and application of engineering and the exploration of its theory, and to solve the following problems:. Firstly, the kinematics model of the ejector boom is established. The ejector arm is a hydraulically driven mechanical arm, which has different driving modes and different kinematic relationships among different joints. The structure of the boom is analyzed in detail. The kinematics model of the arm is obtained, and the driving space of the joint is established. The kinematics relationship between the joint space and the arm terminal task space provides a basis for the motion design and planning from the actuating space of the boom system to the terminal task space. Secondly, the ejector boom consists of seven degrees of freedom, including one redundant degree of freedom, and the inverse kinematics solution is very complicated. An inverse kinematics algorithm which can optimize the position and velocity of joint motion at the same time is designed. The algorithm is mainly based on gradient projection method, and the multi-objective optimization problem is solved by weighted processing of each optimization object. The control goal of preventing the joint displacement from exceeding the limit is achieved on the premise of ensuring the smooth motion of the boom. Finally, because the gradient projection algorithm is a kind of control algorithm based on Jacobian matrix pseudo inverse, so in the process of implementing the algorithm in the computer control system, The inaccuracy of this algorithm is studied in detail, and a closed-loop control strategy is proposed to make the error converge. To sum up, the kinematics model of the boom of jet engine is established, the kinematics problem is discussed in detail, a inverse kinematics algorithm is designed, and a closed-loop control strategy is proposed by improving the control algorithm. The whole design scheme and the research results of the algorithm can provide the theoretical basis and practical reference for the research of the similar mechanical boom.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU643

【参考文献】