Reconfigurable Grasp Planning Pipeline with Grasp Synthesis and Selection Applied to Picking Operations in Aerospace Factories

Several approaches with interesting results have been proposed over the years for robot grasp planning. However, industry suffers from lack of an intuitive and reliable system able to automatically estimate poses while also allowing integration information accumulated knowledge end user. In presented paper it is a non-object-agnostic grasping pipeline motivated by picking use cases aerospace industry. The planning extends functionality simulated annealing optimization algorithm its application within industrial case. Therefore, this addresses first step design reconfigurable modular pipeline. key idea creation functional framework being used factory floor operators according task demands. This software capable generating solutions in offline phase, later on, operation can choose best pose taking into consideration set heuristics that try achieve successful requiring least effort robotic arm. are real environment, relying on mobile platform developed intralogistic tasks. With architecture, new state-of-art methodologies be integrated future growing make more robust applicable wider range cases.