移动机械臂路径规划及全参数运动学算法研究

发布时间：2018-04-10 08:37

本文选题：移动机械臂　切入点：路径规划　出处：《上海应用技术大学》2017年硕士论文

【摘要】：移动机械臂的核心主要由两部分组成,一部分是上部的机械臂部件,另一部分是底部的移动底盘。移动机械臂既保留了机械臂灵活性的同时,又具有可以自由移动的底盘,极大的拓展了机械臂在空间的工作范围。所以,在工业制造、国防、交通运输、仓储物流、服务行业等领域,移动机械臂都具有良好的应用前景。随着我国电子商务、网络购物的迅猛发展,对仓储、物流的自动化及高效率提出了更多的要求,而移动机械臂对仓储、物流行业正是一种绝佳的选择。所以,无论是在科研还是在现实应用等领域,深入开展对其研究都具有重要意义。本文首先对移动机械臂的总体结构进行了介绍,并对各个系统组成部分的特点和需求进行了介绍。归纳了移动机械臂的一些研究热点问题。其次针对目前机械臂的运动学算法普遍只能输入少数几个D-H连杆参数,从而不可避免的存在运动学模型误差的问题,一种拥有更高精度的运动学算法被提出,该算法具有一定的通用性,对具有球形关节的6自由度机械臂适用,可以实现D-H模型中11个D-H参数的输入。为了消除“局部最小点锁死”这个在人工势场法中不可避免的问题,一种融合了常用的栅格地图算法、人工势场思想的算法被提出,该算法不仅可以避免“局部最小点锁死”的问题,而且还能使得算法在计算过程中的数据运算量降低。为了验证算法的有效性,本文通过ROS (开源机器人操作系统),对基于人工势场思想的栅格地图路径规划算法进行了运动学仿真研究。实验研究结果表明,本文提出的基于人工势场思想的栅格地图法能够快速、可靠的规划出最优路径。通过使用Robotics-Toolbox工具箱对本文提出的机械臂的全参数运动学算法进行验证。结果表明,相比传统的运动学算法,机械臂的全参数运动学算法具有更高的模型精度,在机械臂运动学算法中常见的静态误差也可以被很大程度的降低。最后,不仅在ROS仿真环境中实现了移动机器人路径规划算法的仿真,更通过搭建的移动机器人平台,将ROS系统框架移植到了真实的移动机器人平台中,并实现了真实环境中最优路径的生成。
[Abstract]:The core of the mobile manipulator consists of two parts, one is the upper arm and the other is the moving chassis at the bottom.The mobile manipulator not only retains the flexibility of the manipulator, but also has the chassis which can move freely, which greatly expands the working range of the manipulator in space.Therefore, mobile manipulator has a good application prospect in industrial manufacturing, national defense, transportation, warehousing and logistics, service industry and so on.With the rapid development of electronic commerce and online shopping in China, more demands have been put forward for the automation and high efficiency of warehousing and logistics.Therefore, whether in scientific research or in the field of practical applications, it is of great significance to carry out in-depth research on it.In this paper, the overall structure of mobile manipulator is introduced, and the characteristics and requirements of each system component are introduced.Some research hotspots of mobile manipulator are summarized.Secondly, a kinematics algorithm with higher precision is proposed to solve the problem of kinematics model error, which can only input a few D-H link parameters.The algorithm is universal and suitable for a 6-DOF manipulator with spherical joints. The input of 11 D-H parameters in D-H model can be realized.In order to eliminate the inevitable problem of "lock-in of local minimum points" in the artificial potential field method, a raster map algorithm is proposed, which combines the commonly used raster map algorithm and the artificial potential field idea.The algorithm can not only avoid the problem of "local minimum locking", but also reduce the amount of data operation in the calculation process.In order to verify the validity of the algorithm, the kinematics simulation of raster map path planning algorithm based on artificial potential field is carried out by ROS (open source robot operating system).The experimental results show that the raster map method proposed in this paper based on the idea of artificial potential field can plan the optimal path quickly and reliably.The full parameter kinematics algorithm of the manipulator presented in this paper is verified by using the Robotics-Toolbox toolbox.The results show that, compared with the traditional kinematics algorithm, the full parameter kinematics algorithm of the manipulator has higher model accuracy, and the static error commonly used in the kinematics algorithm of the manipulator can be greatly reduced.Finally, not only the path planning algorithm of mobile robot is simulated in the ROS simulation environment, but also the ROS system framework is transplanted to the real mobile robot platform through the mobile robot platform.The generation of optimal path in real environment is realized.
【学位授予单位】：上海应用技术大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP241

【相似文献】