A surrounding point set approach for path planning in unknown environments

Abstract

Due to uncertainties caused by not having full information about the map, path planning in unknown environments is considered to be a more complex problem than in known environments. Consequently, appropriate decisions have to be made at each movement, as the trace depends on which parts of the map have been recognized, and how paths have been locally generated. Therefore, the present paper develops a mechanism for path planning in unknown environments by finding points that surround the importantly considered obstacles. The proposed mechanism can efficiently generate paths within the recognized region, since only subsets of obstacles and points are considered for planning a path, rather than taking all discovered obstacles and all grid points into account; also, in the proposed mechanism, only the newly discovered points by a movement are subject to being processed for updating the path. In experiments, the path planning from the start to the goal was simulated by recognizing the map based on a sensor with a radius. Then, a comparative experiment based on a well-known existing method was used to evaluate the performance of the proposed method. The results demonstrate that the proposed approach can achieve significant efficiency, as well as successfully reach the goal based on the partial information without wandering the map.