Based on observation point fusion, an algorithm named immune network exploration algorithm was proposed to fast accomplish the exploration task of multi-robot for unknown environment. In the case of greatly reducing communication and well calculating the general observation point costs, local observation points of individual robots fused on a map, which can incarnated the collaboration ability of robots enough. Furthermore, the T-cell function was used to update the immune network concentration model, considering that the system performance was affected by diffuse degree and exploring direction of observation points. Simulation results validated that the complete exploration was effectively realized. Simulation results proved that the robots collaboration extent was obviously improved. In addition, the exploration efficiency was greatly improved by adding diffuse degree and exploring direction coincidence indicator.
