SLAM-Assisted Coverage Path Planning for Indoor LiDAR Mapping Systems
This work addresses robust navigation for autonomous indoor mapping systems, presenting an incremental improvement by combining and modifying existing strategies.
The paper tackles the problem of integrating offline coverage path planning with online SLAM for LiDAR-based indoor mapping, resulting in improved total area coverage through modifications to path planning and SLAM error evaluation.
Applications involving autonomous navigation and planning of mobile agents can benefit greatly by employing online Simultaneous Localization and Mapping (SLAM) techniques, however, their proper implementation still warrants an efficient amalgamation with any offline path planning method that may be used for the particular application. In this paper, such a case of amalgamation is considered for a LiDAR-based indoor mapping system which presents itself as a 2D coverage path planning problem implemented along with online SLAM. This paper shows how classic offline Coverage Path Planning (CPP) can be altered for use with online SLAM by proposing two modifications: (i) performing convex decomposition of the polygonal coverage area to allow for an arbitrary choice of an initial point while still tracing the shortest coverage path and (ii) using a new approach to stitch together the different cells within the polygonal area to form a continuous coverage path. Furthermore, an alteration to the SLAM operation to suit the coverage path planning strategy is also made that evaluates navigation errors in terms of an area coverage cost function. The implementation results show how the combination of the two modified offline and online planning strategies allow for an improvement in the total area coverage by the mapping system - the modification thus presents an approach for modifying offline and online navigation strategies for robust operation.