Auction aggregation protocols for agent-based task assignment in multi-hop wireless sensor and robot networks

In this article, several auction aggregation protocols are used for task assignment in multi-hop wireless sensor and robot networks. In the most of existing task assignment solutions (some localized exceptions exists) communication costs are rarely considered. To further reduce communication overhead we propose introduction of agents in the following way. Area monitored by sensors is divided into sectors where one static agent, having all information about robots in that sector, is located in the center. We consider two scenarios. In the first scenario, event information is reported to one of the robots by sensors and robot forwards information to the agent in that sector. In the second scenario event is reported to the agent in that sector by sensors. Having received event information, agent triggers auction searching for the robot with the lowest cost among neighboring agents. Introducing agents that use auction aggregation protocol decreases number of exchanged messages (at least twice compared to experiments without agents) for achieving task assignment decision thus lowering energy consumption. Performances of protocols are shown by simulation results.