Marauli, Tobias (Johannes Kepler University Linz), Gratt, Florian (Johannes Kepler University Linz), Gattringer, Hubert (Johannes Kepler University Linz), Mueller, Andreas (Johannes Kepler University Linz)

Nearly Time-Optimal Online Pick and Place Trajectory Computation Using Quadratic Programming

Scheduled for presentation during the Regular Session "Robot Task Planning" (FrAT2), Friday, July 18, 2025, 10:20−10:40, Room 106

Joint 10th IFAC Symposium on Mechatronic Systems and 14th Symposium on Robotics, July 15-18, 2025, Paris, France

This information is tentative and subject to change. Compiled on August 2, 2025

Abstract

Many robotic applications require fast planning of highly dynamic movements for their execution in time critical scenarios. A successful execution depends on quickly finding trajectories which can accomplish the task. Computing optimal trajectories in a short time is further challenging. The paper addresses this problem in the context of Pick and Place operations. This is formulated as a finite-dimensional time-optimal control problem. Two sets of reference solutions are computed for a discretized workspace using a multiple shooting approach. One set is for picking objects from a moving conveyor and the other one for placing them in a storage. Utilizing a QP-based method enables replanning an admissible reference solution to meet prior unknown picking pose in real-time. The proposed strategy is implemented on an industrial PC controlling a Comau-Racer-3. Measurements show the capability of calculating nearly optimal trajectories online in Pick and Place operations on a moving conveyor.