Z.M. Ma, F. Kabanza, K. Belghith, and R. Nkambou (Canada)
Robot path planning, dangerzones, desirezones.
This paper discusses the problem of planning a robot path in a workplace that contains a set of obstacles, a set of dangerzones, and a set of desirezones. The path cannot collide with the obstacles, should avoid the dangerzones, and should go through the desirezones. In order to plan such a path, we introduce the notion of fuzzy probabilistic roadmap (fuzzy PRM) as an extension of the traditional PRM. Each edge in a fuzzy PRM is assigned a possibility degree to represent the possibility that it is feasible. We also introduce an algorithm for computing a fuzzy PRM that uses lazy collision checking; that is, the collision of edges in the fuzzy roadmap with obstacles is delayed until a candidate path has been computed.
Important Links:
Go Back
