Design approach and simulation of a peristaltic sortation machine

This paper presents the mechatronic design, simulation and data-based control of a peristaltic sortation and singulation machine (PSM) used to operate in parcel distribution centers. Peristaltic movement is a well-known technique from nature for good transport and locomotion. For sortation and singulation, peristaltic movements provide the advantage of gentle and versatile transportation compared to conventional methods. We propose a simulation integrated design procedure which incorporates a discrete element method simulation model early in the development process. This allows for early feasibility studies which speeds up the development process. Due to the need for a novel design approach from scratch, the design is proved using a Finite Element Analysis. A main challenge of the PSM design lies in the development of a suitable control system. As a model-based design appears to be hardly possible due to the complex process which is difficult to model physically, we propose a data-based control design based on reinforcement learning (RL), which reuses the developed discrete element method simulation in a co-simulation framework for the generation of the data. The contributed design, simulation and control approach show the feasibility of a PSM.