Abstract
Robot exploration lies at the heart of fully autonomous system. When extended to the multi-robot case, further time efficiency can be achieved at the cost of an increased complexity of the problem to deal with. Our approach to design such a system does not presuppose the availability of maps of the environment. We first present a 3D landmark and laser based on-line map building associated with a methodology to fuse team robots maps in the general case where no relative poses are given. Details are also given about our 3D landmark recognition system.
