FAKULTÄT FÜR

INFORMATIK

Path planning for industrial robots among multiple under-specified tasks

Industrial robots are flexible and powerful machines. Equipped with the right tool, they can be applied to almost every production task. Robots are very expensive machines, therefore the faster they do their work, the more income they provide. In order to increase efficiency of the industrial robots, different approaches were developed: techniques for deriving robot paths from CAD models, collision-free path planning, automated sequencing of subtasks, etc. This PhD thesis focuses on developing of the algorithms, which optimize the task sequences automatically. In contrast (and synergy) to other approaches, we want to make use of the obvious but generic fact that most robotic tasks allow a certain degree of flexibility but not treated in such a way. Several examples of the task execution flexibility are: closed-contour tasks like cutting or deburring could be started/finished at any point, robot end-effector might approach the starting point with a set of possible orientations, etc. In this PhD project we are looking for such flexibilities and involve them into the task sequence optimization.