A stochastic-computational scheme is presented for generating as-is graph structure spanning complex local geographics. In this schematics, GPS tracks uploaded by a multitude of probe vehicles are exploited to expand a network of feasible roadway patterns. By associating the GPS tracks with a multitude of Brownian motions conditioned by landmark allocation, the graph structure is successively augmented in terms of feasible roadway segments. In reference to the augmented information, symbolic design of over-the-horizon maneuvering process can be interactively substantiated into a chain of topological minors grounded on the local geographics.
Introductory RemarksDespite rapid advancement of vehicle information systems for localizing [11], planning[3], regulating [12] and operating[10] maneuvering processes, the scope of decision making in naturally complex scenes is still confined by on-vehicle control systems to be finally grounded on physical-geometric perspective. Recently, it has been demonstrated that vehicle control systems can be networked and deployed as a multitude of remote-visions for over-the-horizon coordination of the maneuvering processes[8]; images of distant scenes are localized and analyzed by probe vehicles to augment geographic annotations to a cut of satellite image. Such augmented annotations provide logical basis for multi-participant cooperation over the physicalgeometric perspective [7]. For instance, future visitors to the scene can retrieve the annotations to anticipatively adapt the on-board vision systems prior to physical access. To this end, a symbolic plan should be designed in reference to the landmark allocation annotated by probe vehicles as illustrated in Fig. 1 where a future visitor is required to compile a set of landmarks to specify a path connecting the origin and a possible destination. To substantiate the symbolic plan, in this schematics, a multitude of GPS tracks are convened from probe vehicles and matched with the landmark connection within the visualization of the local terrain. Through the symbol-to-trajectory association, a multitude of anticipative and/or graphbased road following systems[6,1] can be collectively organized towards an in-situ instantiation of vehicle * Manuscript Received