【作者】 张斌；

【导师】 张铁；

【作者基本信息】 华南理工大学 ， 机械工程， 2018， 硕士

【摘要】 工业机器人现已广泛应用于砂带打磨作业中,但仍主要采用“示教-再现”的模式生成机器人打磨轨迹,一旦曲面的形状复杂,示教编程不但耗时长而且会降低机器人打磨的质量。为了能够快速地生成高质量的机器人砂带打磨轨迹,针对离线编程模式生成机器人砂带打磨轨迹的方法,对刀路规划、机器人打磨轨迹规划以及机器人打磨轨迹的优化三个方面进行了研究。为了生成高质量的曲面打磨刀位点,对曲面打磨刀路的生成方法与插补方法进行了研究。为解决在曲线曲率变化剧烈处,曲率估算插补算法生成的刀位点数目过于稀疏的缺陷,提出了一种优化的曲线插补算法,该算法在曲线的任何地方都能够生成符合设定误差的刀位点数目,避免了过切的发生。为解决用偏置线法生成等残高刀路所产生的的边界问题和自交问题,提出了一种基于参数线法生成等残高行切刀路的算法,相对于传统的用偏置线法来生成等残高刀路,该算法生成的刀路更稳定,而且没有边界问题和自交问题。仿真结果表明,生成的打磨刀位点具有较高的质量,能够符合打磨加工要求。为了将曲面打磨刀位点转化为机器人的打磨运动轨迹,对机器人打磨轨迹的规划方法进行了研究。为了提高机器人打磨轨迹规划算法的通用性,提出了运用运动螺旋理论对机器人进行正运动学建模,以及运用数值法来求解机器人的逆解,相对于通过D-H参数法建立的机器人正、逆运动学方程,该建模方法对于不同型号的机器人都具有统一的求解形式,实现了通用的机器人正、逆运动学求解算法。为了生成机器人的打磨轨迹,阐述了根据刀位点文件生成机器人打磨轨迹的原理。仿真结果表明,该建模方法具有较好的通用性,而且用上述建模方法生成的机器人打磨轨迹与用D-H参数建模生成的机器人打磨轨迹一致。上述机器人打磨轨迹规划算法得到的机器人轨迹,可能存在关节量变化较大以及打磨过程发生碰撞等缺陷;为解决该问题,对机器人打磨轨迹的生成过程进行了优化,提出了一种机器人优化打磨轨迹搜索算法。首先,提出了可以通过调整砂带轮上打磨点的位置,对机器人的打磨轨迹进行优化;接着,推导并整合了打磨点坐标系调整公式,根据打磨点坐标系的调整公式,提出了用两个坐标轴来描述机器人打磨轨迹,建立轨迹结点的数学模型;最后,提出了一种机器人打磨轨迹搜索算法,得到一条无碰撞且关节量变化较小的机器人打磨轨迹。用V形工件对算法进行仿真,对比优化前机器人打磨轨迹规划算法所得到的机器人轨迹,利用优化算法得到的机器人轨迹的最大关节变化量从2.8123rad减小到1.0741rad;用S形工件对算法进行仿真,表明利用优化算法能够生成一条无碰撞的机器人打磨轨迹,实现了机器人无碰撞砂带打磨。更多还原

【Abstract】 Industrial robots are now widely used in abrasive belt grinding operations,but the mode of teaching and reproduction is still mainly used to produce robot grinding trajectory.Once the shape of the surface is complex,the teaching programming is not only time-consuming but also reduces the quality of robot grinding.In order to using off-line programming mode to fastly produce high quality robot grinding trajectory,the tool path planning,robot grinding trajectory planning and robot grinding trajectory optimization are studied.In order to generate high-quality surface grinding tool locations,surface grinding tool path generation method and interpolation method were studied.In order to solve the problem of sharp curvature of the curve,the number of tool locations generated by the algorithm of curvature estimation interpolation is too sparse,an optimized algorithm of curve interpolation is proposed.The number of the tool locations generated by this algorithm would conform to the setting error in any place of the curve,avoiding the occurrence of over cutting.In order to solve the boundary problems and self-intersection problems generated by offset line iso-scallop tool path generation method,an iso-scallop path algorithm based on parametric line method is proposed,compared with the traditional offset line method,the algorithm is more stable,and there is no boundary problem and self-intersection problem.The simulation results show that the tool locations generated by the algorithm above has high quality and can meet the requirements of grinding.In order to transform the surface grinding tool locations into the robot grinding trajectory,the robot grinding trajectory planning method is studied.In order to improve the universality of the robot grinding trajectory planning algorithm,the screw theory is used to model the positive kinematics of the robot,and the numerical method is used to solve the robot inverse kinematics problem,compared with the forward and inverse kinematics equations established by the D-H parameter method,the modeling method has a unified solution for different types of robots,and a universal algorithm for solving the forward and inverse kinematics of the robot is realized.Then,the principle of generating the robot grinding trajectory based on tool location file is expounded.The simulation results show that the modeling method has good generality,and the robot grinding trajectory generated by the above modeling method is consistent with the grinding trajectory of the robot modeled by D-H parameters.The robot trajectory,generated by the above robot grinding trajectory planning method,may have some defects such as larger joint volume and collision in grinding process.To solve this problem,an optimizing robot grinding trajectory searching algorithm is proposed.First,it is proposed that we can adjust the grinding path of the robot by adjusting the location of the grinding point on the belt wheel.Then,the adjustment formula of grinding point coordinate system is derived and integrated;according to the adjustment formula of the grinding point coordinate system,establishing a mathematical model that using two coordinate axes to describe the robot grinding trajectory.Finally,a robot searching algorithm is proposed to get a collision free robot with small joint size.The algorithm is simulated with V-shaped workpiece,compare and optimize the robot trajectory obtained from the pre robot polishing trajectory planning algorithm,the maximum joint change of robot trajectory obtained from optimization algorithm is reduced from 2.8123 rad to 1.0741 rad.Then the S-shaped workpiece is used to simulate the algorithm,which shows that a collision free robot grinding path can be generated by the optimization algorithm,and the collision free robot belt grinding can be achieved.更多还原