COMPUTATIONAL AND EXPERIMENTAL DETERMINATION OF MANIPULATING ROBOTS RIGIDITY IN DRILLING

Abstract

Manipulating robots are still not widely used in machining, including drilling. The major reason for this is relatively low rigidity of robots. Low rigidity results in low accuracy of the workpiece machined surfaces, including the accuracy of hole axes position. In this regard, the computational prediction of possible drill shift in machining using robots is an urgent problem of mechanical engineering. The most accurate method for prefabricated structures rigidity calculation is the finite element method. However, the consequences of robot finite element meshes generation based on robot CAD models are the meshes consisting of several million elements. Computers currently used by technologists cannot perform such cumbersome calculations. Consequently, the problem arises to perform such calculations using adequate smaller-sized meshes. The paper proposes a method for mesh generation based on voxel modeling, namely, on obtaining a cloud of voxel nodal points using robot CAD model. Robot voxel model was obtained using four assumptions, including: replacing the robot assembly with the equivalent part; the use of hinges equivalent in rigidity obtained from experiments; invariability of the position of the robot links during drilling process. Based on the model regular finite element meshes of up to one million elements were generated. The verification of this calculation method was carried out for displacements of the end link of a three-axis robot. The deviations of the calculated and experimental data do not exceed 18%. Taking into account many difficult-to-formalize calculation factors, for example, cutting forces, this deviation is acceptable for predicting the accuracy the hole axis location in drilling with the use of robots. Consequently, the proposed calculation method is promising for the pre-production engineering.