Method, robot and multi agent system for robot operation in multi agent system

摘要

The present invention is a method for operating a multi-agent system having a plurality of robots (1), and each robot (1) has the following methods until the target system state is reached, that is, the current method. Starting from the system state (q1-q10), a plurality of possible options are determined in which the progression along the path of the system state in a predetermined deterministic finite automaton (DFA) is achieved (S11), where the options are: It defines an action that allows a transition from the current system state to the subsequent system state (q1-q10) to be achieved: • For each of the multiple possible options, the action specified by that option. The cost value to be executed is determined (S12), and the auction is executed (S14, S15). At that time, the cost value determined for each option is considered by the remaining robot (1). The present invention relates to a method of periodically executing an action corresponding to one of a plurality of options depending on all cost values determined or received for the corresponding option (S16).