【作者】 李发忠；

【导师】 赵德安；

【作者基本信息】 江苏大学 ， 农业电气化与自动化， 2012， 博士

【摘要】 静电喷涂机器人是一种非常重要的涂装生产装备,是一种利用高压静电电场力提高涂料微粒沉积效率的设备,在国内外广泛应用于汽车车身涂装生产线。(静电喷雾机器人在农田的静电喷药杀虫和除草等方面也有越来越多的应用研究)。静电喷涂机器人喷枪轨迹对喷涂对象的表面喷涂质量影响很大,因此,静电喷涂机器人喷枪轨迹优化算法、控制策略和离线编程系统的研究是国内外学者们近几年关注的热点。我国在喷涂机器人喷涂轨迹优化方面的研究起步较晚,并且研究的深度和广度也不及国外,喷涂机器人及其离线编程系统的技术水平较低,在实际应用中的效果不太理想。针对以上问题,本文在国家自然科学基金项目“静电喷涂机器人变量喷涂轨迹优化关键技术研究(60875052)”和江苏省工业高技术研究项目“喷涂机器人轨迹优化与离线编程技术研究(BG2006023)”的资助下,对喷涂机器人轨迹优化与离线编程系统中的喷枪位姿信息的获取、轨迹优化和工件空间轨迹到机器人空间轨迹的转换等关键技术问题进行了研究。主要研究内容如下：1.针对外形复杂的曲面工件难以获得准确的喷枪轨迹,影响涂层质量的问题,提出一种基于点云切片技术的喷涂机器人喷枪位姿参数的自动获取方法。首先,对待喷涂工件进行3D扫描直接获得工件表面点云模型,然后,转换为三角化网格格式(STL)模型,利用STL模型切片技术由程序自动生成喷涂机器人的喷枪运行轨迹。通过该方法可以方便地获取喷枪轨迹的位置和姿态参数,该参数可用于机器人离线编程系统中的喷涂路径规划,从而实现对喷枪的运动轨迹、方向以及喷枪与待喷涂工件表面之间距离的精确控制。在此基础上对已有轨迹进行优化可以获得更好的喷涂效果。2.针对在自由曲面上的匀速和等行程间距喷涂时,会因曲面上曲率的变化造成涂层厚度差较大的问题,提出一种基于速度和间距的轨迹优化算法以提高曲面的涂膜厚度一致性。根据平面上的喷涂实验和微分几何的面积放大定理推导出曲面上的漆膜生长模型；采用Gauss-Bonnet定理选择曲面上的测地线作为种子路径曲线；采用沿路径方向的速率优化算法补偿表面曲率变化对沿轨迹方向的涂层一致性的影响；根据已知种子曲线的位置,通过优化间距获得其他路径的位置,使临近路径的涂料沉积产生适当的重叠,补偿表面曲率变化对垂直于路径方向的涂层厚度一致性的影响；建立速度优化和间距优化的多目标优化函数,对整片工件上的轨迹进行综合优化,得到整体最优轨迹。以汽车车身表面为实验对象,进行清漆喷涂实验验证了算法的有效性。3.针对采用固定参数的喷涂轨迹在有些情况下会造成涂料的严重浪费和非规则平面工件边界处涂膜厚度不一致的问题,提出一种基于静电电压和间距调节的变量喷涂轨迹优化方法。首先,研究静电电压、旋杯转速、间距和平移速率等工况参数对高压静电旋杯式喷涂机器人的喷涂模型和喷涂转移率的影响,根据喷涂时静电电压与喷涂模型之间关系建立了基于高压静电旋杯的变量喷涂模型。然后,根据变量喷涂模型的特点研究了规则平面和非规则平面的变量喷涂轨迹优化方法。在不规则多边形平面上进行喷涂时,通过改变静电电压、移动速度、轨迹间的间距等参数控制涂料的分布规律,从而改变涂料沉积图形的大小以获得最优的喷涂效果和效率。最后,通过复杂非规则平面多边形的仿真和实验验证了变量喷涂轨迹优化方法的可行性。4.针对目前轨迹优化是基于工件空间的喷涂轨迹优化,没有涉及机器人运动轨迹这一问题,根据喷涂机器人离线编程的需要和喷涂空间轨迹的特点,提出一种基于Dijkstra算法的喷涂机器人笛卡尔空间轨迹到关节轨迹的转换方法。通过分析工件坐标系和机器人基坐标系的关系,采用辅助特征点三点标定法将工件坐标系内的喷涂空间轨迹转换到机器人坐标系。建立在机器人逆解中求取最短关节运动行程的优选模型,利用赋权有向图表示所有机器人逆解间的行程关系,利用Dijkstra算法求最优的逆解组合。喷涂试验验证了该方法的可行性和有效性,结果表明所提出的轨迹转换方法能有效克服轨迹转换失真。围绕变量喷涂轨迹优化技术进行以上四部分的理论分析和试验研究工作,基本构建了喷涂机器人离线编程系统的基本框架,为我国的静电喷涂机器人轨迹优化技术的发展奠定了部分理论基础。更多还原

【Abstract】 Electrostatic spray robot is an important equipment for coating, which has been widely applied in coating production line of automobile at home and abroad. Since it can improves the deposition efficiency of charged paint particles via high electrostatic field force, the research on trajectory optimization algorithm, control strategy and off-line programming system of electrostatic spray robot has attracted tremendous interest in recent years. The research of trajectory optimization of spray robot began very late in China and research fields are very limited. The technical level of spray robot and off-line programming system is low, which makes the electrostatic spray robot unfavorable for real applications. This paper will investigates some practical problems in development of spray painting robot. This work is supported in part by the National Natural Science Foundation Program (Grant No.60875052)—"Research on Trajectory Optimization of Electrostatic Spray Robot with Variable Parameter" and in part by the High-tech Research Program of Jiangsu (Grant No.BG2006023)—"Research on Trajectory Optimization and Off-line Programming Technology of Spray Robot". The main content and contributions of the thesis are listed as follows:1. Due to the complex geometry of free-form surfaces, it is still a challenge to generate optimization trajectories of spray gun that satisfies paint uniformity requirements. A method of acquiring the position and orientation parameters of spray painting robot automatically via point cloud slicing technology is proposed. The trajectory of spray gun is usually composed of a set of equidistant curves, which is similar to the cross-section contours created by uniform slicing. The point cloud data of the work-piece is obtained by 3D laser scanning, then, slicing process is applied to the STL model in order to obtain the intersection points. After the slice direction and number of slices are defined by user, the intersection algorithm is adopted to generate the slice profiles of point cloud model. The spraying position is determined by the average sampling of cross-section contours, and the normal vectors of all the sampling points are estimated by using the weighted average method. Finally, the trajectory of spray gun is generated by offsetting the sampling points along their normal vectors. Experimental results illustrate that the method owns good feasibility and effectiveness. The reason is that the spraying trajectory, direction and distance of spray gun can be controlled accurately.2. Due to the big coating thickness difference on free-form surfaces, caused by the change of surface curvature, a method of spray robot trajectory optimization based on the optimization of speed and distance is provided to improve the uniformity. A quadratic function of the paint deposition rate on a plane is proposed according to the experimental data, and a model of paint deposition rate on a free-form surface is also established. In order to select the relative position of the seed curve on the surface, we employ the Gauss-Bonnet theorem. An algorithm of variable spray speed optimization is established to compensate for the curvature change and improve the uniformity of paint deposition along passes. An algorithm for selecting the spacing between the passes is proposed to improve the deposition uniformity vertical to passes, as well as reduce cycle time and paint waste. Finally, the experiment on automobile demonstrates the feasibility and availability of these optimization algorithms.3. Due to the big coating thickness difference and waste on the boundary of irregular polygon plane when the parameters of spray trajectories are constant, a method of trajectory optimization of electrostatic spray robot with variable parameter based on the electrostatic voltage and distance is provided. We first investigate the effect of spray model and transfer efficiency by the parameters of static electricity voltage, rotate speed of ESRB, velocity of ESRB and space between bell and work-piece etc, an then proposed a variable parameter spray model of high-voltage electrostatic rotating bell according to the relationship between the electrostatic voltage and spray model. In order to obtain the optimum spraying effectiveness and efficiency when spraying of irregular polygon plane, we control the distribution of paint and changing the size of the paint deposition graphics by changing electrostatic voltage, moving speed, the distance between passes, etc. Simulations and experiments on the complex non-regular planar polygon demonstrate the feasibility of this trajectory optimization method.4. A conversion method from cartesian space trajectory to joint trajectory of spray painting robot, based on Dijkstra algorithm, is presented to meet the requirements of offline programming spray painting robot and characteristics of spray cartesian space trajectory. Through analyzing the relation between the part coordinates and the robot basal coordinates, the spray cartesian space trajectory is converted to the robot basal coordinates by three points calibration of auxiliary feature points. An optimized model is then proposed to obtain the shortest joint motion distance in robot inverse kinematics solutions. Moreover, the distance among all of the inverse kinematics solutions is described by a directed weighted graph, and the Dijkstra algorithm is employed to obtain the optimal combination of inverse kinematics solution. The feasibility and efficiency of the method are well verified by spray experiments.Based on the technology of trajectory optimization with variable parameter,the theoretical analysis and experiments in the above four sections are derivedfor electrostatic spray robot, which builds the basic frame of spray robot’s off-lineprogramming system. Meanwhile, the research will partly establish the theoretical basis for thedevelopment of the electrostatic spray robot.更多还原