No abstract

Abstract-This paper presents a new method for constructing 2D maps of enclosed underwater structures using an underwater robot equipped with only a 2D scanning sonar, compass and depth sensor. In particular, no motion model or odometry is used. To accomplish this, a two step offline SLAM method is applied to a set of stationary sonar scans. In the first step, the change in position of the robot between each consecutive pair of stationary sonar scans is estimated using a particle filter. This set of pair wise relative scan positions is used to create an estimate of each scan's position within a global coordinate frame using a weighted least squares fit that optimizes consistency between the relative positions of the entire set of scans. In the second step of the method, scans and their estimated positions act as inputs to a mapping algorithm that constructs 2D octree-based evidence grid maps of the site. This work is motivated by a multi-year archeological project that aims to construct maps of ancient water storage systems, i.e. cisterns, on the islands of Malta and Gozo. Cisterns, wells, and water galleries within fortresses, churches and homes operated as water storage systems as far back as 2000 B.C. Using a Remotely Operated Vehicle (ROV) these water storage systems located around the islands were explored while collecting video, still images, sonar, depth, and compass measurements. Data gathered from 3 different expeditions has produced maps of over 60 sites. Presented are results from applying the new mapping method to both a swimming pool of known size and to several of the previously unexplored water storage systems.

Abstract. We present a methodology and algorithm for the reconstruction of three dimensional geometric models of ancient Maltese water storage systems, i.e. cisterns, from sonar data. This project was conducted as a part of a four week expedition on the islands of Malta and Gozo. During this expedition, investigators used underwater robot systems capable of mapping ancient underwater cisterns and tunnels. The mapping included probabilistic algorithms for constructing the maps of the sonar data and computer graphics for surface reconstruction and visualization. This paper presents the general methodology for the data acquisition and the novel application of algorithms from computer graphics for surface reconstruction to this new data setting. In addition to reconstructing the geometry of the cisterns, the visualization system includes methods to enhance the understanding of the data by visualizing water level and texture detail either through the application of real image data via projective textures or by more standard texture mapping techniques. The resulting surface reconstructions and visualizations can be used by archaeologists for educational purposes and to help understand the shape and history of such water receptacles.

Abstract:Geometric data acquired via a scanning process can suffer from holes due to errors in the acquisition process, noise, or challenges in merging multiple inputs together into a unified map. We present a straight forward algorithm to fill holes in incomplete evidence grids representing acquired geometric data. We also present our methods to apply learning in order to statistically evaluate the proposed hole filling algorithm. This analysis validates our proposed method for hole filling and additionally enables the construction of a probability distribution function to represent the accuracy of the filled data per model. During surface reconstruction, this function can be used to visualize the certainty of the filled geometry via transparency and coloring giving the user an understanding of the data's accuracy. This work is motivated by a multi-year project to construct educational visualizations of ancient water storage systems, i.e. cisterns and wells within churches, fortresses and homes on the islands of Malta, Gozo and Sicily.