| The production of aluminum alloy wheels has been basically automated,but the grinding process is still manual.Therefore,robotic grinding has gradually attracted the attention from academia and industry,and researchers mainly focus on the design of the grinding actuator and the research of the grinding mechanism.As an important part of robot grinding,path planning is one of the determinants of grinding effect and efficiency and needs more research.This dissertation takes force-controlled robotic grinding of the wheel hub surface as the object and studies its path planning algorithm.The input is three-dimensional point cloud of the upper surface of the hub.The grinding and polishing tool is an elastic disk and its control strategy is force-control.This dissertation divides complex surface into idea sub surfaces through segmentation algorithm based on sweeping lines,plan the local path in each sub surface introducing grinding mechanism,optimize the non-grinding path and finally,analyze and verify the above algorithm through simulation and manufacturing process experiments.The main work is listed as follows:Aiming at the problem that the direct path planning on the complex surface will cause the path to be interrupted and make the grinding tool leave the surface of the workpiece,this paper proposes a surface segmentation algorithm based on sweeping lines,so that any path composed of parallel sweeping lines is guaranteed be continuous in any sub-surface.Morse decomposition based on specified function pattern is a common algorithm to do the same job in full coverage path planning.Inspired by this,this paper proposes a surface segmentation algorithm based on sweeping lines.First,digitizes the function pattern,and then clusters the sweeping lines by judging the overlap of adjacent sweeping lines.Each cluster represents one final sub-surface.Aiming at the multi-objective local path planning problem of simultaneously optimizing the grinding effect and efficiency,this paper proposes a greedy algorithm by continuously solving constrained optimization problems and a global algorithm by solving single large-scale optimization problem to plan the feed rate and path spacing.First,according to the grinding mechanics model,the material removal rate per unit time on the surface of the workpiece is calculated through the grinding force and related parameters.Then,establish a constrained nonlinear optimization problem with the goal of maximizing the total grinding volume based on discretized single grinding path points and solve it to plan the feed rate of this grinding path.Finally,based on the feed rate planning model and results,the greedy algorithm is used to plan the distance from the next grinding path;or discretize multiple grinding path,and optimize the multiple path spacings and corresponding feed rates at the same time with the goal of minimizing the average residual volume to be polished.Aiming at the problem of grinding interval movement planning,that is,the problem of non-grinding path planning,this paper constructs an integer programming problem that takes the result of solving the traveling salesman problem as the objective function,and then optimize the types of local paths and the order of visiting all the sub-surfaces with the goal of minimizing the total length of interval movement. |
