Establishing cost surface model is of basis for conducting path optimisation in continuous space. In most cases, path optimisation calculation in continuous space has been based on a single resolution cost surface model; however, some problems could be triggered at the same time, viz. data redundancy, high computational costs, susceptible to “edge effect” of ground features and etc., when it is adopted to solve routing optimisation problems as regards the constructions of power lines, pipelines, roads, railways and so forth. With a purpose upon solving above problems, this paper proposes a variable resolution raster cost surface model for path optimisation, and elaborates its design ideas and modelling method. The experimental results show that this model not only can model the ground feature density and terrain complexity effectively, but it also solves the problems aroused by the single resolution model. After a comparative analysis on the calculation results of this model and the traditional single resolution cost surface model, findings show that the proposed model can get reasonable paths in varied environments with high computational efficiency.
