Research On Robot Trajectory Optimization For Regular Surfaces Grinding

As one of the high-precision machining methods,surface grinding has a decisive influence on the surface quality of workpiece.Grinding trajectory step control method is one of the effective ways to ensure high-precision and high-efficiency machining,but the chord-high error method in the step control method does not guarantee the maximum machining step length within the chord tolerance range.At the same time,the strategy of trajectory optimization ignores the influence of robot speed and vibration on the smoothness of grinding trajectory and surface processing quality in the actual processing application.Based on the above situation,this paper researches on the processing trajectory of the combined surface composed of regular surfaces,such as plane,cylinder,torus,etc.The method of equal chord height error step is improved,the maximum machining step in each feed is ensured and the efficiency of searching step point is improved.The smoothness of grinding trajectory is generated by combining speed,acceleration and vibration in the actual processing scene.An optimization scheme is proposed to improve the curved surface quality of grinding surface.The main work is as follows:(1)This work introduces the influence of grinding mode and grinding parameters on the processing quality of grinding surface.And then it analyses the geometric characteristics of the surface,derives the homogeneous expression of surface,calculates the curvature of the surface and judges the concavity and convexity of surface,which provides theoretical support for the grinding trajectory generation algorithm.(2)In order to solve the problem of misjudgment and inefficiency in searching the next processing point by the equal chord height error step method,this paper improves the equal chord height error step method based on the quasi-Newton DFP algorithm,combines methods of the midpoint and three-point calibration and uses the super-linear convergence of the DFP algorithm to search quickly for the maximum step length in the range of the pitch tolerance.The method reduces the number of tool and improves the efficiency of algorithm.(3)Based on the equal chord height error step method improved by DFP algorithm,the grinding trajectory generation module is realized.The experiment results of offline simulation and actual machining of grinding surface prove the effectiveness of the grinding trajectory generation module.(4)With respect to the problem of poor smoothness of grinding trajectory in the actual processing application,this paper applies the robot motion command program derived from offline programming software to the actual processing scene.This work collects robot motion data and analyses velocity error,acceleration continuity and vibration.The optimal velocity relation based on parameter fitting and the dynamic control parameter optimization scheme are proposed to reduce the vibration in machining and improve the smoothness of machining track.