A rule-based enhancement of a vision guided, collision-free robotic application

Abstract

This paper presents a rule-based method to solve a collision issue for an industrial robot manipulator. The approach applies a set of rules over the information received from a vision system to consider an extended set of constraints for end-of-arm tooling. Collision-free, finding objects that are safe for gripping or the order of operation are based on a graph dependency mechanism. The interaction among a robot, a vision system, and the developed module is obtained through an IoT system extension. Validated on a real manufacturing system, our method can simplify and improve several robotized applications.