Many efforts have been done in correcting the ability of a robot to follow curved trajectories while being accurate in the repetitive positioning at specific points. CutRob+ is focused in this problematic; the project will implement a formal model for evaluating the performance of real path tracking systems and recommend the most suitable corrected trace. The model will provide a way to grade the tracking performance of a particular robot based on an interactive learning process while considering the configuration of specific parameters that could influence in the final quality. In order to accomplish this objective the solution is supported by a knowledge based expert system and a specially designed computer vision system that will be interoperable among different protocols by using open interfaces. While the proposed solution can be generalized to any robot arm control, special care will be placed on Plasma Arc Cutting systems (PACs) pilot. Further improvements will be also studied in the field of online image processing for PACs
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement nº 721267.