Engineering intuition for designing multi-robot search and rescue solutions

here has been il lot of interest over the last decade in the use of multiple robots in applications such as search and rescue missions. The use of multiple robots could increase mission effectiveness and robustness. The design of algorithms for robot cooperation to achieve collective mission goals is still on-going research. Various possibilities depending also on the resources available, namely the communication range and bandwidth, computational power and memory, have been previously proposed. As in most system engineering problems, the design of multiple robot solutions will be driven partly by intuition. This paper is an initial attempt to provide a multiple robot applications designer some engineering intuition for the solution development. For a specific multiple robot search and rescue application, three solutions over a spectrum of possibilities are compared. The three solutions range from very limited to more capable communications and processing. The light weight approaches are based on Swarm Intelligence which mimics insect behaviour and the heavy weight approach is based on Multi- Agent Systems which mimic human behaviour. For each solution, the performance is derived through numerous simulations. The various performances are then compared. Attempts are also made to identify key characteristics of each solution and to relate them across solutions and to the simulation results.