Decentral task allocation for industrial AGV-systems with routing constraints

Abstract

Automated guided vehicles (AGVs) form a large and important part of the logistics transportation systems in today's industry and are widely used. One of the main problems in the control of AGV systems is the routing problem, where all robots attempt to execute all of their allocated tasks without congestions with other robots. This is a complex problem and a large variety of solutions exist in literature to this purpose. One possible solution that is not widely investigated in literature is the possibility of including routing information as a constraint in the task allocation process in order to obtain a more balanced allocation that could reduce possible congestions. In task allocation literature, costs for a robot to execute a task are largely estimated without considering possible delays that can occur due to congestions when moving to a task. In this paper, the observed research gap of including routing constraints in the task allocation problem is addressed by proposing a decentralized task allocation algorithm based on sequential single-item (SSI) auctions with the implementation of delegate-MAS (DMAS) as a routing constraint in the bidding process. Our new approach is benchmarked to an SSI-solver that does not take routing constraints into account.