در تحقیقات تجربی رفتار ساختاری روبات ها

In the paper, the structural behavior of industrial robots is investigated. The objective is the development of a model, capable of predicting the robot's accuracy, under certain arm positions and loading conditions. The Finite Element Method (FEM) is used for the model's development. An extended investigation into the total robot accuracy of the joint effect is conducted. The accuracy of the robot, under ranging loads at different positions, has been mapped and discussed.

In order for the manufacturing industry to increase its production flexibility, it has been proposed that the milling machines be substituted with robots [1]. However, the accuracy and repeatability (+-0.07mm) of a typical robotic arm [2] is not as high as this of an ordinary CNC milling machine with typical value of ±0.016 mm or better [3]. Since the accuracy issues affect the quality of the final product, they have to be resolved in order for the penetration of machining robots to be increased in industry. This paper focuses on the modeling of a robotic arm, in FE environment. The purpose is to simulate its behavior under various loading scenarios and create a visual representation of the robot’s performance in its working envelop, in terms of accuracy. Developments in machining and tool design technology, especially in milling operations, reflect the requirements for flexibility in order to adapt the changes taking place in the market and in the global economic environment [1]. Makhanov et al. [4] presented a new approach to tool-path optimization of milling robots, based on a global interpolation of the required surface by a virtual surface composed of tool trajectories. Kao et al. [5] presented a robot-based computer-integrated manufacturing (CIM) automation. Abele et al. [6] described the modeling of the robot structure and the identification of its parameters focusing on the analysis of the system’s stiffness and its behavior during the milling process.

The accuracy of the original model in comparison with that of the experimental data has shown extended deviations. In order for these deviations to be confronted in the new model created, the changes were incorporated into the joint properties to approach the actual robot behavior. The model was counterpoised with experimental data and through adjustments of the joint stiffness value, the model was fine tuned. Further experiments, on the actual robot arm, showed that the model followed the actual behavior of the robotic arm, with minimum deviations. To better depict the model accuracy, a deflection map has been created, clearly showing the simulated accuracy of the model in the workspace of the robot arm.