Communication Aspects of Coordination in Robot Wireless Networks

obots coordinate among themselves to selectone of them to respond to an event reportedto one of the robots so that the communi-cation cost of selecting the best robot,response time, and cost of performing thetaskareminimized.Existingsolutionsareeithercen-tralized, assuming a complete graph, or based onfloodingwithindividualresponsestoarobotdeci-sion maker [simple auction protocol (SAP)],ignoring communication cost and response timebound. This article proposes auction aggregationprotocols (AAPs) for task assignment in multihopwireless robot networks. A robot collector leadsan auction and initiates a response tree construc-tion by transmitting the search message. Afterreceiving the message, each robot makes a deci-sion on whether to retransmit a search message,based on the estimated response cost of its robots,up to k-hops away. Robots wait to receive thebids from its children in the search tree. Then,robots aggregate responses by selecting the bestbidandforwarditbacktowardtherobotcollector(auctioning robot). When distance is used as thesole cost metrics, the traversal aggregation algo-rithm [routing with face traversal (RFT) —routingtoward the event with the traversal of the face con-tainingtheevent]canbeappliedandisanoptimalsolution. Several other protocols and theirenhancementsarealsodescribedhere.Multirobot systems (MRSs) are wellstudied in literature [1], and the focalpoint of the majority of MRS-relatedarticles is on coordination and coop-eration. The term networked roboticsemerged recently emphasizing that robotscan be connected by a wireless mediumforminga communication network.Thereare