Global path planning in 4D environments using topological mapping

Global path planning is a challenging problem raised in many fields of research. It is of particular interest to construction planning community facing the requirements of trustworthiness and feasibility of the project schedules. Correct construction schedules must exclude conflicting situations at project sites, thereby improving project productivity and reducing risks and waste. Some conflicts can be foreseen and avoided using emerging 4D modeling and planning technologies. Modern 4D modeling systems enable to identify and report simple clashes among construction elements, equipment units, workspaces occupying the same place at the same time. However, they are not capable of revealing more sophisticated conflicts caused by lack of collision-free paths to deliver the elements to final destination locations. Global path planning methods look promising to assure the existence of collision-free paths for displaced elements. In particular, topological mapping methods have proved to be well suited for single path planning requests in static 3D scenes. However, they could not be directly applied to complex pseudo-dynamic scenes originating from 4D modeling and planning applications. In this paper a new efficient method intended for global path planning in dynamic environments is presented. It generalizes existing mapping techniques by means of incremental and concordant updates of all the deployed metric and topological structures as the explored environment is being evolved under events appearing in discrete time moments. The obtained results, both theoretical and experimental, showed that the proposed method gives a significant performance gain for multiple path planning requests in complex pseudo-dynamic environments and can be successfully applied for validation of construction project schedules against path conflicts.