C. Chen and H. Wang
SLAM, loop-closing, active contour, optimization
This paper presents a mapping method that, with a single robot visiting on area just once, can accurately map as large environment. The resulting map provides a thorough 3D description of the environment in a predefined global coordinate system. Our first contribution is to represent the map as a collection of submaps arranged in a deformable configuration and to perform loop-closing by registering this submap configuration to an aerial image. The second contribution is to introduce the active contour technique to the SLAM domain, so that the registration is efficiently solved in an iterative energy minimization process. The constraints from robot mapping are modeled as forces trying to keep the submaps consistent, while the pictorial matching is represented by forces guiding submaps to a configuration consistent with an aerial image. In the experiment, we demonstrate the proposed algorithm’s capability to close a 1,890 meter loop with only one visit. The result is compared with ground truth and high accuracy is observed.
Important Links:
Go Back
