一种支撑式管道机器人运动控制系统的设计含开题及5张cad图
收藏
资源目录
压缩包内文档预览：
编号：145707983
类型：共享资源
大小：4.33MB
格式：ZIP
上传时间：2021-09-27
上传人：QQ14****9609
认证信息
个人认证
郭**（实名认证）
陕西
IP属地：陕西
50
积分
- 关 键 词：
-
一种
支撑
管道
机器人
运动
控制系统
设计
开题
cad
- 资源描述：
-
一种支撑式管道机器人运动控制系统的设计含开题及5张cad图,一种,支撑,管道,机器人,运动,控制系统,设计,开题,cad
- 内容简介：
-
译文：一种移动机器人运动规划系统穆拉特伊尔迪里姆摘要:本文提出了一种移动机器人运动规划系统。路径规划是在有障碍物的环境中，为移动机器人从起点节点移动到目标节点寻找一条可行的路径。遗传算法利用其较强的寻优能力来生成最优路径。通过摄像机和图像处理实现了移动机器人、障碍物和目标的定位。为运动规划系统设计了一个图形用户界面(GUI)，允许用户与机器人系统交互，并观察机器人环境。所有系统的软件组件是用MATLAB编写的,它提供了使用设定配件而不是机器人固件,以避免混乱的c+库机器人的专有软件,详细控制机器人,而不是编译程序中经常实时动态操作。 1.介绍近年来，随着技术的发展，研究人员对自动驾驶汽车的兴趣越来越大。文献中有很多关于自动驾驶汽车的研究。路径规划是自动驾驶汽车研究的主要课题之一。路径规划是在障碍物1的环境中，寻找移动机器人从起点节点移动到目标节点的可行路径。路径规划环境可以是静态的，也可以是动态的。在静态环境中，必须在开始执行之前找到整个解决方案。然而，对于动态或部分可观察的环境，每一步都需要频繁地重新规划，这需要更多的规划更新时间为了解决移动机器人的路径规划问题，已经发展了许多方法。每种方法的有效性取决于应用程序环境的类型，每种方法都有自己的优缺点。与基于微积分和枚举策略等传统的搜索和优化方法相比，进化算法具有鲁棒性、全局性，而且通常更容易应用于对问题只有很少或没有先验知识的情况下求解2。在过去的十年中，遗传算法利用其较强的寻优能力3被广泛应用于生成最优路径。遗传算法已被公认为最鲁棒搜索技术之一的复杂和不良行为的目标函数。遗传算法在开发接近最优解方面具有吸引力的基本特征是它们本质上是并行搜索技术4,5。他们可以同时以并行的方式搜索所有的工作环境，从而更快地找到更好的解决方案。先锋P3-DX移动机器人是一种流行的研究机器人，可编程，适合教室和实验室使用。很多研究人员在他们的研究中都使用了这个机器人6,7,8,9。ARIA是面向对象的机器人控制应用程序接口，用于移动机器人的智能机器人。使用ARIA的程序员应该熟悉使用典型的c+概念，包括使用简单继承的类和对象、指针、内存管理、STL容器以及编译和链接进程10。然而，MATLAB，这是一个流行的学术软件，是一种高级语言和交互式环境执行计算密集型任务比传统的编程语言，如C, c+和FORTRAN更快。使用MATLAB，程序员可以比使用传统语言更快地编写和开发算法，因为不需要执行低级管理任务，如声明变量、指定数据类型和分配内存。此外，MATLAB每次只执行一个命令或一组命令，而不编译和链接11。本文介绍了一种移动机器人的运动规划系统。利用摄像机和数字图像处理技术对移动机器人自身、障碍物和目标进行定位。先锋P3-DX移动机器人用于室内应用。采用遗传算法进行无碰撞路径规划。基于网格的环境模型作为移动机器人的运动区域，在室内应用中经常用到。为运动规划系统设计了一个图形用户界面(GUI)，允许用户与机器人系统交互，并观察机器人环境。包括图像处理、路径规划以及与机器人和GUI的通信，系统的所有软件组件都是用MATLAB编写的，而不是使用ARIA的c+库。在MATLAB编程中提供了使用非预定义的附件而不是机器人的固件，避免了在c+库中混淆，详细控制机器人和不重新编译程序频繁的实时动态操作。 2运动规划系统ARIA是一个面向对象的机器人控制应用程序，它有一个编程库(SDK)，供希望访问他们的P3-DX平台和附件的c+程序员使用。然而，机器人固件不执行任何高级机器人任务。相反，它是运行在连接PC上的智能客户端来执行这种应用级机器人控制策略和任务的工作，如障碍物检测和躲避、定位、地图、智能导航和摄像头控制12。移动机器人室内运动规划涉及定位、路径规划和运动控制等问题。图1为本研究所使用的运动规划系统。定位是确定移动机器人、障碍物和目标的位置。通常位置由用户给出或由摄像机识别。在实时动态工作的情况下，如在环境中使用移动的障碍物或目标，应首选通过摄像机进行定位。图1所示：提出了一种移动机器人运动规划系统由于本研究采用了实时和动态应用，因此采用了摄像机和图像处理技术进行定位。摄像头安装在天花板上，它将环境的实时图像发送给计算机。每个物体上都有不同颜色的标签;绿色表示障碍物，蓝色表示目标，黄色表示移动机器人。为了确定移动机器人的航向角，也用红色标记。机器人上黄色和红色圆心的连线方向等于机器人的航向角。在计算机上运行MATLAB编写的图像处理程序，从相机获取瞬时图像，如图2左上角所示。它能识别物体上的颜色和底座的颜色。这一过程对光线的均匀性和环境中的反射很敏感。该程序确定每个对象的坐标，其颜色与基本颜色不同，如图右上图所示。坐标的确定是一个复杂的问题，因为真实环境和它的图像不会由于摄像机的方向而重叠。因此，需要利用真实环境和图像的角坐标进行坐标变换。最后，图像处理程序将颜色坐标与基于网格的地图进行匹配，如图右下所示。这样就实现了本地化，生成了地图。如图1所示的运动规划系统的第二步是将地图发送到路径规划过程。本研究采用遗传算法来确定移动机器人所经过的路径。使用遗传算法的路径规划将在下一节中详细说明。最后，为了控制机器人平台，一个用MATLAB编写的客户端程序通过机器人连接发送命令包。这可以使用直接命令来完成。直接命令由两个字节的数据包头、一个字节的字节数、一个字节的命令数、一个字节的参数类型、n个字节的参数和两个字节的校验和组成，由机器人的操作系统ARCOS13定义。直接命令方法允许直接向机器人平台发送任何不寻常或特殊的命令，而不需要任何干预处理。图2：利用图像处理手段制作地图 3利用遗传算法进行路径规划遗传算法是一种模拟自然遗传算子的并行全局搜索技术。因为它同时评估参数空间中的许多点，它更有可能收敛到全局最优。搜索空间不一定是可微的或连续的。遗传算法将自然选择机制启发的算子应用于编码参数空间的基因串群体中。在每一代中，它探索搜索空间的不同区域，然后将搜索导向有较高可能性找到更好的解决方案的区域。A .环境和染色体的表示许多路径规划方法都采用基于网格的模型来表示环境空间。研究表明，基于网格表示的障碍物距离计算和障碍物表示更容易。基于网格的环境空间有两种表示方式，一种是坐标平面4,14,15，另一种是有序编号网格1,16,17。坐标可以用二进制数或十进制数表示。图3给出了路径规划环境的二进制、十进制和有序编号网格表示。环境上的阴影网格表示不可行的障碍物区域，空白网格表示移动机器人可以自由移动的可行区域。染色体表示路径规划问题的候选解18。一条染色体或一条路径由起始节点、目标节点和移动机器人通过的跳跃节点组成。这些路径上的节点或步骤称为染色体的基因。不同的编码方法用于创建染色体(图4)，这取决于环境的表示方法。一般采用二进制编码串法4,19,20，但也采用十进制编码串法1,16,17，被认为更灵活。十进制编码需要更少的时间和空间计算开销。图3。环境的二进制、十进制和有序编号表示图4。染色体的二进制、十进制和有序编码B .种群初始化初始总体一般是随机产生的。一些产生的染色体可能包括与障碍物相交的不可行的路径。即使初始种群中包含不可行路径，遗传算子也能找到最优或接近最优解。然而，这一过程降低了算法的搜索能力，增加了寻找最优解的时间。此外，两个不可行染色体的交叉可能产生新的不可行路径。为了解决这一问题，在产生初始群体时，必须检查每条染色体是否相交于一个障碍。如果是，则用可行基因随机改变染色体的相交基因。确定了以可行初始种群作为遗传算法的起始点是相当有利的，如16,17所述。C.适合度函数和选择 路径规划问题的目的是找到从起始节点到目标节点之间的最优路径。最优路径可能是最短的21，在它上绊倒所需时间和能量最少的路径。在路径规划问题中，一般将目标函数视为最短路径。移动机器人从前一个节点移动到下一个节点后，其x-y坐标和航向角的变化如图5所示。在本研究中，遗传算法中染色体的目标函数值如式(1)(2)所示。图5.移动机器人位置的改变在那里;F是适应度函数，pk是染色体的第k个基因(节点)，n是染色体的长度，d是两个节点之间的距离，xk和yk是机器人当前的水平和垂直位置，xk+1和yk+1是机器人下一个水平和垂直位置。机器人的方向由式(3)确定。目标函数值定义为路径中各节点之间的距离之和。如果在机器人的方向上有障碍物，则在目标函数值上加一个惩罚。惩罚值应该大于环境中的最大路径长度。为了找到最优路径，该算法搜索一个染色体的惩罚被消除。国籍：美国出处：AECE-电气和计算机工程的进步原文：A Motion Planning System for Mobile RobotsMurat YildirimAbstract：In this paper, a motion planning system for a mobile robot is proposed. Path planning tries to find a feasible path for mobile robots to move from a starting node to a target node in an environment with obstacles. A genetic algorithm is used to generate an optimal path by taking the advantage of its strong optimization ability. Mobile robot, obstacle and target localizations are realized by means of camera and image processing. A graphical user interface (GUI) is designed for the motion planning system that allows the user to interact with the robot system and to observe the robot environment. All the software components of the system are written in MATLAB that provides to use non-predefined accessories rather than the robot firmware has, to avoid confusing in C+ libraries of robots proprietary software, to control the robot in detail and not to re-compile the programs frequently in real-time dynamic operations. I. INTRODUCTIONRecently, interest of researchers on autonomous vehicles increases with technological developments. There are many studies in the literature about autonomous vehicles. One of the main subjects on autonomous vehicle is path planning. Path planning tries to find a feasible path for mobile robots to move from a starting node to a target node in an environment with obstacles 1.The path planning environment can be either static or dynamic. In the static environment, the whole solution must be found before starting execution. However, for the dynamic or partially observable environments re-planning are required frequently in each step and it needs more planning update time.There are so many methods that have been developed to overcome the path planning problem for mobile robots. Each method differs in their effectiveness depending on the type of application environment and each one of them has its own strengths and weaknesses. Compared to traditional search and optimization methods, such as calculus-based and enumerative strategies, the evolutionary algorithms are robust, global and generally more straightforward to apply in situations where there is little or no prior knowledge about the problem to solve 2.In the last decade, genetic algorithms have been widely used to generate the optimal path by taking the advantage of its strong optimization ability 3. Genetic algorithms have been recognized as one of the most robust search techniques for complex and ill-behaved objective functions. The basic characteristic that makes the GA attractive in developing near-optimal solutions is that they are inherently parallel search techniques 4, 5. They can search all working environment simultaneously in a parallel manner and so they can reach a better solution more quickly.The Pioneer P3-DX of Mobile Robots is a popular research robot that is programmable and suitable for classroom and laboratory use. So many researchers have used this robot in their study 6,7,8,9. ARIA is an objectoriented robot control application-programming interface for intelligent robots of Mobile Robots. Programmers working with ARIA should be familiar with using typical C+ concepts, including using classes and objects with simple inheritance, pointers, memory management, the STL containers, and the compiling and linking process 10. However, MATLAB, which is the one of popular academic software, is a high-level language and interactive environment to perform computationally intensive tasks faster than with traditional programming languages such as C, C+, and FORTRAN. With the MATLAB, a programmer can program and develop algorithms faster than with traditional languages because there is no need to perform low-level administrative tasks, such as declaring variables, specifying data types, and allocating memory. Moreover, MATLAB executes commands or groups of commands one at a time, without compiling and linking 11.In this study, a motion planning system for a mobile robot is introduced. A camera and digital image processing is used for localization of the mobile robot itself, obstacles and the target. The Pioneer P3-DX of Mobile Robots is used for indoor applications. A genetic algorithm is used for collision-free path planning. Grid-based environment model, which is frequently used in indoor applications, is used as the motion area of mobile robot. A graphical user interface (GUI) is designed for the motion planning system that allows the user to interact with the robot system and to observe the robot environment. Including the image processing, path planning, and communication with the robot and the GUI, all the software components of the system is written in MATLAB instead of using C+ libraries of ARIA. Programming in MATLAB provides to use non-predefined accessories rather than the robot firmware has, to avoid confusing in C+ libraries, to control the robot in detail and not to re-compile the programs frequently in real-time dynamic operations.II. MOTION PLANNING SYSTEMARIA is an object-oriented robot control application that has a programming library (SDK) for C+ programmers who want to access their P3-DX platform and accessories. However, the robot firmware does not perform any highlevel robotic tasks. Rather, it is the job of an intelligent client running on a connected PC to perform this application-level robotic control strategies and tasks, such as obstacle detection and avoidance, localization, mapping, intelligent navigation and camera control 12.Indoor motion planning of mobile robot deals with the issues of localization, path planning and the motion control. Fig. 1 shows the motion planning system used in this study. Localization is the determination of the positions of mobile robot, obstacles and the target. Generally positions are given by a user or identified by a camera. In the case of real-time and dynamic working, such as moving obstacles or targets are used in the environment, localization by means of a camera should be preferred.Figure 1. Proposed motion planning system for mobile robotBecause real-time and dynamic applications are made in this study, a camera and image processing techniques are used for localization. The camera is mounted on the ceiling and it sends the real-time images of the environment to a computer. Each object is labeled with a different color on it; green for the obstacles, blue for the target and yellow for the mobile robot. In order to determine the heading angle of the mobile robot, it is also labeled with a red color. Direction of the line that connects the centers of yellow and red colored circles on the robot equals to the heading angle of it.Image processing program running on the computer, which is written in MATLAB, takes the instantaneous images from the camera, as shown on the upper left image of Fig. 2. It identifies the colors on the objects and the color of the base. This process is sensitive to the homogeneity of light and the reflections in the environment. The program determines the coordinates of each object which has a color different from the base color, as shown in the upper right image of the figure. Coordinate determination is a complicated problem, because the real environment and its image do not overlap due to the direction of the camera. Therefore, coordinate transformation should be done by using corner coordinates of the real environment and the image. At the last, image processing program matches the coordinates of colors with a grid based map, as shown on the lower right image. In this way, localization is realized and the map is produced.The second step of the motion planning system shown in Fig. 1 is to send the map to the path planning process. In this study, genetic algorithm is used to determine the path which mobile robot goes through it. Path planning with the genetic algorithm is explained in detail in the following section.In the last step, in order to control the robot platform, a client program, which is written in MATLAB, sends command packets through the robot connection. This can be done using direct commands. Direct commands consist of two-byte packet header, one-byte byte count, one-byte command number, one-byte argument type, n-byte argument and two-byte checksum, as defined by the robots operating system ARCOS 13. The direct command method allows sending any unusual or special command directly to the robot platform, without any intervening processing.Figure 2. Map production by means of image processing III. PATH PLANNING WITH GENETIC ALGORITHMSGA is a parallel and global search technique that emulates natural genetic operators. Because it simultaneously evaluates many points in the parameter space, it is more likely to converge toward the global optimum. It is not necessary that the search space to be differentiable or continuous. GA applies operators inspired by the mechanics of natural selection in a population of gene string which is encoding the parameter space. At each generation, it explores different areas of the search space, and then directs the search to regions where there is a high probability of finding a better solution.A. Representation of Environment and ChromosomeMany path planning methods use a grid-based model to represent the environment space. It has been determined that calculation of distance and representation of obstacle is easier with grid-based representation. The grid-based environment space is represented in two ways, by the way of coordinate plane 4,14,15 or by the way of orderly numbered grids 1,16,17. Coordinates can be represented with both the binary or decimal numbers. Fig. 3 shows the binary, decimal and orderly numbered grid representation of the path planning environment. The shadowed grid on the environment shows the infeasible obstacle area and blank grid shows the feasible area where mobile robots can move freely.A chromosome represents a candidate solution 18 for the path planning problem. A chromosome or a path consists of a starting node, a target node and the hopping nodes which mobile robot across to them. These nodes or steps in the path are called as genes of the chromosome. Different coding methods are used to create chromosomes (Fig. 4), depending on the representation method of the environment. Binary coded string method 4,19,20 is used in general, however decimal coded string method is also used 1,16,17 and it is thought as to be more flexible. Decimal coding needs less computational overhead in time and space. Figure 3. Binary, decimal and orderly numbered representations of the environmentFigure 4. Binary, decimal and orderly numbered coding of chromosomesB. Initialization of PopulationThe initial population is generally generated randomly. Some of the generated chromosomes may include infeasible paths which intersect an obstacle. An optimal or near optimal solution can be found by genetic operators, even though the initial population includes infeasible paths. However, this process reduces the search capability of the algorithm and increases the time to find an optimal solution. Furthermore, crossover of two infeasible chromosomes may generate new infeasible paths. In order to solve this problem, each chromosome must be checked whether it intersects an obstacle, when generating the in
- 温馨提示:
1: 本站所有资源如无特殊说明，都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
2: 本站的文档不包含任何第三方提供的附件图纸等，如果需要附件，请联系上传者。文件的所有权益归上传用户所有。
3.本站RAR压缩包中若带图纸，网页内容里面会有图纸预览，若没有图纸预览就没有图纸。
4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
5. 人人文库网仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对用户上传分享的文档内容本身不做任何修改或编辑，并不能对任何下载内容负责。
6. 下载文件中如有侵权或不适当内容，请与我们联系，我们立即纠正。
7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。
人人文库网所有资源均是用户自行上传分享，仅供网友学习交流，未经上传用户书面授权，请勿作他用。
川公网安备: 51019002004831号