Realisation of a following function for mobile autonomous robots

Abstract

Autonomous mobile robots (AMR) are steadily finding their way into production and logistics halls, but also gradually for the transport of goods on the last mile in urban environments. In addition to the autonomous navigation of these systems, the use of a following function has the potential to increase the efficiency in intralogistics and order picking, as well as to enable alternative modes of individual transport in urban environments. The AMR detects and identifies the user, keeps a defined distance to them and actively avoids obstacles. In the following, an AMR will be presented that realises this following function through a model predictive control approach (MPC) and probabilistic estimation methods for movement prediction.