Accurate prediction of interception positions for catching thrown objects in production systems

Abstract

This work aims to optimize transportation processes in production by throwing objects between the working stations instead of transporting them on conveyed belts. One aspect of this new approach is the accurate prediction of interception positions of thrown objects with the catching robot. In a first step, this work analyzes the variation of the flight trajectories to specify the requirements for an appropriate catching-device. Several objects were thrown by a throwing-device and the range of their passages through a gate similar to the portal of the proposed gantry robot for object catching were measured. The second and main part of this work deals with calculating the objectpsilas position in flight and accurately predicting a point of interception with the catching robot. The proposed prediction model for interception positions bases on two inputs: the starting angles extracted from observations of a single camera and the speed of the object measured by two light barriers. The model is precise enough in respect to the size of the gripper of the catching robot.