Self-Configuration of Waypoints for Docking Maneuvers of Flexible Automated Guided Vehicles

We study the problem of automatic configuration of the initial position, the so-called waypoint, from which to initiate a robust and accurate docking maneuver using nonholonomic (car-like) robotic forklifts in the context of automated manufacturing. The proper selection of these positions is of paramount importance to operate with the industrial grade of accuracy, repeatability, and reliability required by load transfer operations in industrial settings. An unconstrained optimization method coupled with probabilistic techniques is proposed to solve this problem. The proposed method permits to increase significantly the flexibility and adaptability of the autonomous robotic forklifts.