Self Organizing Maps for AUVs Mapping Silvia Botelho #1 , Celina da Rocha #2 , Gabriel Oliveira #3 , Mˆ onica Figueiredo #4 , Paulo Drews #5 # Universidade Federal do Rio Grande (FURG) Italia Av. Km 8 Rio Grande, Rio Grande do Sul,Brasil 1 silviacb@furg.br 2 celinahaffele@yahoo.com.br 3 gabrielleivas@gmail.com 4 monica.sfigueiredo@hotmail.com 5 dudopel@gmail.com Abstract— The use of Autonomous Underwater Vehicles (AUVs) for underwater tasks is a promising robotic field. These robots can carry visual inspection cameras. Besides serving the activities of inspection and mapping, the captured images can also be used to aid navigation and localization of the robots. In this context, this paper proposes an approach to mapping of underwater vehicles. Supposing the use of inspection cameras, this proposal is composed of the development of topological maps using self-organizing maps and Growing Cell Structures(GCS) for localization and navigation. A set of tests was accomplished, regarding online and performance issues. The results reveals an accuracy and robust approach to several underwater conditions, as illumination and noise, leading to a promissory and original visual mapping technique. I. I NTRODUCTION Autonomous Underwater Vehicles (AUVs) are mobile robots that can be applied to many tasks of difficult human exploration [1]. In underwater visual inspection, the vehicles can be equipped with down-looking cameras, usually attached to the robot structure [2]. These cameras capture images from the deep of the ocean. In these images, natural landmarks, also called keypoints in this work, can be detected allowing the AUV localization and mapping. In this paper we propose a new approach to AUV mapping. Our approach extract and map keypoints between consecutive images in underwater environment, building online keypoints maps. This maps can be used to robot localization and navi- gation. We use Scale Invariant Feature Transform (SIFT), which is a robust invariant method to keypoints detection [3]. Fur- thermore, these keypoints are used as landmarks in an online topological mapping. We propose the use of self-organizing maps (SOM) based on Kohonen maps [4] and Growing Cell Strutures (GCS) [5] that allow a consistent map construction even in presence of noisy information. First the paper presents a detailed view of the SOM and GCS. Next, our approach is presented, followed by the im- plementation, test analysis and results with different undersea features. Finally, the conclusion of the study and future per- spectives are presented. II. USING SOM AND GCS FOR AUV MAPPING Figure 1 shows an overview of the approach proposed here. First, the underwater image is captured and pre-processed to removal distortions caused by water diffraction [6]. With the corrected image, keypoints are detected and local descriptors for each one of these points are computed by SIFT. Each keypoint has a n dimensional local descriptors and global pose informations. A matching stage provides a set of correlated keypoints between consecutive images. The relative motion between frames is estimated, using the correlated points and the homography matrix [7]. Fig. 1. System Overview. The keypoints are used to create and train the topological maps. The growing cell strutures algorithm is used to create the nodes and edges of the SOM. Each node has a n-dimensional weight. After a trainning stage, the system provides a topo- logical map, where its nodes represent the main keypoints of the environment. During the navigation, when a new image is captured, the system calcules its local descriptors, correlating them with the nodes of the current trainned SOM. Next, it is detailed the proposed approach.