双SCARA机器人无碰路径规划及应用研究
发布时间：2018-07-28 08:53
【摘要】：作为柔性制造系统关键组成部分的机器人已广泛应用于电子制造行业。如何进一步地提升产能是业界关注的重点,为弥补单机器人系统的不足,应用多机器人系统成为一个常用的解决方案,而如何规划多机器人在共享工作空间下的无碰运动则是一个亟需解决的问题。本文以多机器人在电子制造行业中的应用为背景,围绕多机器人无碰路径规划和碰撞检测等关键技术及在异型插件作业中的应用展开研究,主要完成了如下工作:1.多机器人系统无碰路径规划问题的描述以及基本方法研究。探讨了多机器人系统在位形空间中的描述方式,将多机器人系统无碰路径规划问题转化为在位形空间中的路径求解问题;为解决位形空间下的路径求解问题,采用了单次查询、双向采样与延迟碰撞检测相结合的SBL-PRM(Single-query,Bi-directional,Lazy collision checking,Probabilistic Roadmap Method)进行路径规划;设计了多机器人无碰路径规划器的总体方案,为多机器人系统的无碰路径规划问题提供通用解决框架。2.在多机器人无碰路径规划器的框架下设计与实现了双SCARA机器人无碰路径规划器。为达到更好的路径搜索效果,确定了SBL-PRM中的最大采样次数、邻域半径、局部路径检测阈值;针对PRM产生路径的次优性,提出自适应步长路径平滑算法,提高了无碰路径的规划质量;采用五次多项式对无碰路径点进行规划,提高了机器人运动的平稳性。3.双SCARA机器人PCB异型插件工作站的开发。讨论了工作站的系统组成,完成了工作站控制系统的总体设计,分析了控制系统的控制流程,并对双SCARA无碰路径生成器、逻辑控制器、仿真平台、机器人通讯等关键子程序进行了详细设计和分析。4.实现了多种情景下的无碰路径规划及运动仿真,并将算法进一步应用于双SCARA机器人PCB异型插件工作站中进行验证。仿真和实验结果表明了所提出的无碰路径规划算法和相应的规划器具有较好的有效性和实用性。本文提出的多机器人无碰路径规划方法可推广到含有工业机器人或者移动机器人的多机器人系统中,所开发的双SCARA机器人PCB异型插件工作站对电子行业产能升级具有较大的参考价值和实用价值。
[Abstract]:As a key component of flexible manufacturing system, robot has been widely used in electronic manufacturing industry. In order to make up for the shortage of single robot system, the application of multi-robot system has become a common solution. How to plan the collision-free motion of multi-robots in shared workspace is an urgent problem. Based on the application of multi-robots in electronic manufacturing industry, this paper focuses on the key technologies of multi-robot collision free path planning and collision detection and its application in irregular plug-in operation. The main work is as follows: 1. Description and basic methods of path planning for multi-robot systems without collision. In this paper, the description of multi-robot system in configuration space is discussed. The path planning problem of multi-robot system without collision is transformed into a path solving problem in configuration space, and a single query is used to solve the path solving problem in configuration space. The SBL-PRM (Single-Queryry Bi-directional SBL-PRM) (Single-Queryry Bi-directional collision checking probabilistic Roadmap Method) combined with bidirectional sampling and delayed collision detection is used for path planning, and the overall scheme of multi-robot no-collision path planner is designed, which provides a general solution framework for multi-robot system's no-collision path planning. Under the framework of multi-robot no-collision path planner, a dual SCARA robot no-collision path planner is designed and implemented. In order to achieve better path search effect, the maximum sampling times, neighborhood radius and local path detection threshold in SBL-PRM are determined, and an adaptive step size path smoothing algorithm is proposed for the sub-optimality of PRM generation path. The planning quality of the collision free path is improved, and the smoothness of robot motion is improved by using the fifth order polynomial to plan the path point without collision. The development of PCB special plug-in workstation for dual SCARA robot. The system composition of the workstation is discussed, the overall design of the workstation control system is completed, the control flow of the control system is analyzed, and the dual SCARA collision-free path generator, logic controller and simulation platform are discussed. The key subprograms such as robot communication are designed and analyzed in detail. 4. The collision free path planning and motion simulation are realized in various scenarios, and the algorithm is further applied to the PCB special-plug-in workstation of dual SCARA robot for verification. The simulation and experimental results show that the proposed path planning algorithm and the corresponding planning instruments are effective and practical. The multi-robot path planning method proposed in this paper can be extended to multi-robot systems with industrial robots or mobile robots. The developed PCB plug-in workstation of dual SCARA robot has great reference value and practical value for capacity upgrading in electronic industry.
【学位授予单位】：广东工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242
本文编号：2149571
[Abstract]:As a key component of flexible manufacturing system, robot has been widely used in electronic manufacturing industry. In order to make up for the shortage of single robot system, the application of multi-robot system has become a common solution. How to plan the collision-free motion of multi-robots in shared workspace is an urgent problem. Based on the application of multi-robots in electronic manufacturing industry, this paper focuses on the key technologies of multi-robot collision free path planning and collision detection and its application in irregular plug-in operation. The main work is as follows: 1. Description and basic methods of path planning for multi-robot systems without collision. In this paper, the description of multi-robot system in configuration space is discussed. The path planning problem of multi-robot system without collision is transformed into a path solving problem in configuration space, and a single query is used to solve the path solving problem in configuration space. The SBL-PRM (Single-Queryry Bi-directional SBL-PRM) (Single-Queryry Bi-directional collision checking probabilistic Roadmap Method) combined with bidirectional sampling and delayed collision detection is used for path planning, and the overall scheme of multi-robot no-collision path planner is designed, which provides a general solution framework for multi-robot system's no-collision path planning. Under the framework of multi-robot no-collision path planner, a dual SCARA robot no-collision path planner is designed and implemented. In order to achieve better path search effect, the maximum sampling times, neighborhood radius and local path detection threshold in SBL-PRM are determined, and an adaptive step size path smoothing algorithm is proposed for the sub-optimality of PRM generation path. The planning quality of the collision free path is improved, and the smoothness of robot motion is improved by using the fifth order polynomial to plan the path point without collision. The development of PCB special plug-in workstation for dual SCARA robot. The system composition of the workstation is discussed, the overall design of the workstation control system is completed, the control flow of the control system is analyzed, and the dual SCARA collision-free path generator, logic controller and simulation platform are discussed. The key subprograms such as robot communication are designed and analyzed in detail. 4. The collision free path planning and motion simulation are realized in various scenarios, and the algorithm is further applied to the PCB special-plug-in workstation of dual SCARA robot for verification. The simulation and experimental results show that the proposed path planning algorithm and the corresponding planning instruments are effective and practical. The multi-robot path planning method proposed in this paper can be extended to multi-robot systems with industrial robots or mobile robots. The developed PCB plug-in workstation of dual SCARA robot has great reference value and practical value for capacity upgrading in electronic industry.
【学位授予单位】：广东工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242
【参考文献】
相关期刊论文 前6条
1 吴鸿敏;苏满佳;管贻生;陈新;;动态调整操作序列的多机器人无碰运动规划[J];机器人;2016年06期
2 谢碧云;赵京;刘宇;;基于快速扩展随机树的7R机械臂避障达点运动规划[J];机械工程学报;2012年03期
3 李占坤;赵德安;姬伟;张超;;基于工业PC机的果树采摘机器人控制系统设计[J];机械设计与制造;2010年06期
4 丁富强,韩卫军,赵锡芳;双臂机器人无碰撞运动规划[J];机械科学与技术;2002年06期
5 周明,孙树栋,彭炎午;基于遗传算法的多机器人系统集中协调式路径规划[J];航空学报;2000年02期
6 钱东海,马毅潇,赵锡芳;双臂SCARA型机器人C空间快速建立算法[J];机械科学与技术;1999年01期
相关硕士学位论文 前4条
1 张皓钧;SCARA机器人运动学参数与视觉标定研究[D];哈尔滨工业大学;2013年
2 王锋;基于立体视觉与SBL-PRM算法的自主收获机器人路径规划[D];江苏大学;2010年
3 罗枫;三维网格模型的快速碰撞检测及相交体计算[D];浙江大学;2005年
4 郭海儒;基于OBB层次包围盒树的实时碰撞检测算法[D];太原理工大学;2004年
，本文编号：2149571
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2149571.html
