An Integrated Approach for Motion Law Optimization in Partially Compliant Slider-Crank Mechanisms

Servo-Actuated Mechanisms (SAM) are capable of improving the flexibility and reconfigurability of modern automatic machines. On one hand, as compared to fully mechanical drives, SAM may suffer from non-negligible positioning inaccuracies, whose effect can become unacceptable in case of undesired part deformations during high dynamic motions. On the other hand, it may be the case that parts of the system are purposely designed to provide an highly compliant behaviour, so as to potentially increase the device safety in case of interaction with humans. In both cases, practical strategies to reduce the SAM positioning errors are necessary. As a possible solution to such issue, in this paper, an integrated approach to improve the accuracy of a partially compliant SAM in position-controlled tasks is described. The method exploits a multi-software framework comprising Matlab and RecurDyn, namely a commercial Computer-Aided Engineering (CAE) tool that can be used to simulate the motion of systems comprising both rigid and flexible bodies. Starting from an initial, sub-optimal, motion law of the input link, a trajectory optimizer iteratively runs the CAE solver and automatically computes an optimal, compensated, position profile. The obtained results show that the method may represent a useful tool for analyzing/designing partially compliant SAM, whenever analytical models are either too complex or not readily available.