We study the problem of mapping an initially unknown environment with autonomous mobile robots. More precisely, we consider simplistic agents that move from vertex to vertex along the boundary of a polygon and measure angles at each vertex. We show that such agents are already capable of drawing a map of any polygon in the sense that they can infer the exact geometry up to similarity. Often, such tasks require the agent to have some prior bound on the size of the environment. In this paper, we provide an efficient reconstruction algorithm that does not need any a priori knowledge about the total number of vertices.keywordsangle measurementcomputational geometryexploration strategysimple polygonvisibility graphthese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
|Series||Springer Tracts in Advanced Robotics|