MES Specific Data Analysis. Case Study with the Baxter Robot

Abstract

In this research paper we aim to improve the functions of the Baxter robot [1] through data mining methods. Our case study belongs to the robotics domain, integrated in the context of Manufacturing Execution Systems (MES) and Product Lifecycle Management (PLM). The experimental data includes the parameters registered during the activities of the robot, such as the movement of the left or right arm and refers to collision events. The state of the art concerning the data mining methods is described, then our solution is detailed by presenting the approached data mining techniques. The adopted methods are detailed and then the experimental results are presented and discussed. Finally, the conclusions and further development possibilities are formulated, highlighting the utility of the adopted data mining methods, based on our previously stated objectives. An accuracy above 98% was achieved concerning the validation of our model.