Waypoint-Based Robot Navigation Using NURBs-Based Metamodels

Effective path planning techniques for mobile robots have potential uses and ramifications in a wide range of fields. As a result, the study of these techniques is highly diverse, and historically rich. In this vein we present a path planning methodology for a particular type of robot — one which flies above a terrain but must periodically land at waypoints. The approach demonstrated here is notable for drawing its inspiration and working basis from several different engineering fields. NURBs-based metamodels, a concept developed in the field of engineering design, are used to represent terrain to be navigated. Goal programming, largely used in operations research, is used to determine optimal landing waypoints. Graph theory concepts (derived from the field of computer science) are used to generate robot paths, and more broadly as data structures for information management. We demonstrate the new method on a variety of terrain maps, and present our conclusions on the effectiveness of the algorithm. We close with remarks regarding potential future developments and improvements that may be made to increase the utility of this approach.