Josep Aulinas

Visual SLAM for underwater vehicles using video velocity log and natural landmarks

A visual SLAM system has been implemented and optimised for real-time deployment on an AUV equipped with calibrated stereo cameras. The system incorporates a novel approach to landmark description in which landmarks are local sub maps that consist of a cloud of 3D points and their associated SIFT/SURF descriptors. Landmarks are also sparsely distributed which simplifies and accelerates data association and map updates. In addition to landmark-based localisation the system utilises visual odometry to estimate the pose of the vehicle in 6 degrees of freedom by identifying temporal matches between consecutive local sub maps and computing the motion. Both the extended Kalman filter and unscented Kalman filter have been considered for filtering the observations. The output of the filter is also smoothed using the Rauch-Tung-Striebel (RTS) method to obtain a better alignment of the sequence of local sub maps and to deliver a large-scale 3D acquisition of the surveyed area. Synthetic experiments have been performed using a simulation environment in which ray tracing is used to generate synthetic images for the stereo system.

Feature based slam using side-scan salient objects

Different underwater vehicles have been developed in order to explore underwater regions, specially those of difficult access for humans. Autonomous Underwater Vehicles (AUVs) are equipped with on-board sensors, which provide valuable information about the vehicle state and the environment. This information is used to build an approximate map of the area and estimate the position of the vehicle within this map. This is the so called Simultaneous Localization and Mapping (SLAM) problem. In this paper we propose a feature based submapping SLAM approach which uses side-scan salient objects as landmarks for the map building process. The detection of salient features in this environment is a complex task, since sonar images are noisy. We present in this paper an algorithm based on a set of image preprocessing steps and the use of a boosted cascade of Haar-like features to perform the automatic detection in side-scan images. Our experimental results show that the method produces consistent maps, while the vehicle is precisely localized.

Selective Submap Joining for underwater large scale 6-DOF SLAM

Autonomous Underwater Vehicles (AUVs) need positioning systems different than the Global Positioning System (GPS), which does not work in underwater scenarios. A possible solution to this lack of GPS signal are the Simultaneous Localization and Mapping (SLAM) algorithms. SLAM algorithms aim to build a map while simultaneously localize the vehicle within it. These algorithms suffer from several limitations in front of large scale scenarios. For instance, they do not perform consistent maps for large areas, mainly because uncertainties increase with the scenario. In addition, the computational cost increases with the map size. It has been demonstrated that the use of local maps reduces computational cost and improves map consistency. Following this idea, in this paper we propose a new SLAM technique based on using independent local maps, combined with a global level stochastic map. The global level contains the relative transformations between local maps. These local maps are updated once a new loop is detected and the amount of overlapping between local maps is high. Thus, maps sharing a high number of features are updated through fusion, maintaining the correlation between landmarks and vehicle. Experimental results on real data obtained from the REMUS-100 AUV show that our approach is able to obtain large map areas consistently.

Submapping SLAM Based on Acoustic Data from a 6-DOF AUV

Abstract Autonomous Underwater Vehicles (AUVs) need positioning systems besides the Global Positioning System (GPS), since GPS does not work in underwater scenarios. Possible solutions are the Simultaneous Localization and Mapping (SLAM) algorithms. SLAM algorithms aim to build a map while simultaneously localizing the vehicle within this map. However, they offer limited performance when faced with large scale scenarios. For instance, they do not create consistent maps for large areas, mainly because uncertainties increase with the scale of the scenario. In addition, the computational cost increases with the map size. The use of local maps reduces computational cost and improves map consistency. Following this idea, in this paper we propose a new SLAM approach that uses independent local maps together with a global level stochastic map. The global level contains the relative transformations between local maps. These local maps are updated once a new loop is detected. Local maps that are sharing a high number of features are updated through fusion, maintaining the correlation between landmarks and vehicle. Experimental results on real data obtained from the REMUS-100 AUV show that our approach is able to obtain large map areas consistently.

SLAM based Selective Submap Joining for the Victoria Park Dataset

Abstract One of the main drawbacks of current SLAM algorithms is that they do not result in consistent maps of large areas, mainly because the uncertainties increase with the scenario. In addition, as the map size grows the computational costs increase, making SLAM solutions unsuitable for on-line applications. The use of local maps has been demonstrated to be useful in these situations, reducing computational cost and improving map consistency. Following this idea, this paper proposes a technique based on using independent local maps together with a global stochastic map. The global level contains the relative transformations between local maps, which are updated once a new loop is detected. In addition, the information within the local maps is also corrected. Thus, maps sharing a high number of features are updated through fusion and the correlation between landmarks and vehicle is maintained. Results on synthetic data and on the Victoria Park Dataset show that our approach is able to consistently map large areas and the computational costs are lower.