Topology optimization of rigid-links for industrial manipulator considering dynamic loading conditions

In this work, topology optimization of robotic rigid-links is analyzed considering dynamic loading conditions. Usually, in such a scenario, the topologies are generated considering worst-case or static conditions. However, this generated topology will not be the optimum for other angular positions (dynamic-condition), as it is dependent on load-direction rather than load-magnitude. Here, a method is proposed similar to an equivalent static load technique to synthesize a single topology, which performs better in all angular positions. The method consists of superimposition of individual optimal topologies corresponding to different angular positions, followed by normalization and re-penalization to attain the desired volume fraction. Further, morphological variations are performed using image processing techniques to reduce the stress value and geometric complexities. The synthesized topology by this method shows 10-25% reduction in deflection and stress value compared to any of the optimal topologies. The proposed methodology is illustrated on the two rigid-links of an indigenously developed 3-degree-of-freedom industrial manipulator test rig. It uses three AC-servo motors, controlled using Motion perfect-software with programmable logic controller and human-machine interface. The simulated results are validated using the experimentations. A reduction of 30% link-volume minimizes joint torques by 24.9%, with better values of deflection and stress. (C) 2020 Elsevier Ltd. All rights reserved.