Pianificazione del movimento e controllo di un sistema automatico di stoccaggio per macchine a taglio laser

This thesis delves into a novel and scalable approach for motion planning in dual-gantry kinematics. Each gantry system has two arms, and this method effectively addresses the self-collision issue while ensuring G2-continuity and minimizing curvatures. Starting from collision-free start and target positions, the proposed technique defines the geometric master path using classical five-segment planar trajectory, described by fifth degree spline. Using a path scaling technique, the multidimensional spline defining the geometric path of the four end-effectors is then generated from the master path. The rototranslation of the loaded parts is evaluated in frames to establish sampled overlapping curves. These vectors are then used to compute splines that modify previously defined geometric paths to avoid collisions. In this process, a quadratic programming minimization is employed to reduce the curvature of these splines while preserving the kinematic properties of the original path. Isoparametric trajectory planning ensures compliance with kinematic axis constraints, resulting in an optimized motion profile to reduce task execution time. Finally, this approach is applied to a sorting system for a laser machine (LST) with a maximum of fourteen degrees of freedom. Two case studies for a single gantry configuration are presented and discussed to illustrate and clarify the methodology in typical sorting operations. The choice of detailing these case studies is justified as that configuration is the most widely used in this industrial field. The first of these applications involves a sequence of movements in which the anti-collision procedure is not used and corresponds to at least half of the machine’s movements. In fact, these movements correspond to all those performed with unloaded end-effector tools and those in which the loaded part is in the shadow of the arm. Afterwards, some results regarding this case study are proposed. They are obtained by comparing the new method with the existing one in terms of overall performance. A more complex case study, in which the anti-collision algorithm is used, is then examined in depth. It’s worth noting that the approach outlined in this work can be extended to other robotic systems with similar kinematic structures, providing a valuable tool for motion planning in pick-and-place applications.