PolySLAM: A 2D Polygon-based SLAM Algorithm

Dichtl, Johann; Le, Xuan Sang; Lozenguez, Guillaume; Fabresse, Luc; Bouraqadi, Noury

doi:10.1109/icarsc.2019.8733647

Cited by 3 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this method did not implement the global optimization. Dichtl et al [16] proposed a SLAM method that generates maps in the PolyMap format [19]. The construction of polygons from fitted lines and the polygon map merging using the BSP-tree were explored.…”

Section: Vector-based Slammentioning

confidence: 99%

“…However, constructing such a map relies on high-precision positioning and manual labeling. Existing SLAM methods that can produce a vector-based map were all proposed for indoor environments [13][14][15][16]. Strong assumptions of the indoor structures adopted in these studies, such as orthogonal and parallel walls, cannot be applied to road environments.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Vector-Based Road Structure Mapping Using Multibeam LiDAR

Zhao

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

The high-definition map (HD-map) of road structures is crucial for the safe planning and control of autonomous vehicles. However, generating and updating such maps requires intensive manual work. Simultaneous localization and mapping (SLAM) is able to automatically build and update a map of the environment. Nevertheless, there is still a lack of SLAM method for generating vector-based road structure maps. In this paper, we propose a vector-based SLAM method for the road structure mapping using vehicle-mounted multibeam LiDAR. We propose using polylines as the primary mapping element instead of grid maps or point clouds because the vector-based representation is lightweight and precise. We explored the following: (1) the extraction and vectorization of road structures based on multiframe probabilistic fusion; (2) the efficient vector-based matching between frames of road structures; (3) the loop closure and optimization based on the pose-graph; and (4) the global reconstruction of the vector map. One specific road structure, the road boundary, is taken as an example. We applied the proposed mapping method to three road scenes, ranging from hundreds of meters to over ten kilometers and the results are automatically generated vector-based road boundary maps. The average absolute pose error of the trajectory in the mapping is 1.83 m without the aid of high-precision GPS.

show abstract

Section: Vector-based Slammentioning

confidence: 99%