At 110m in diameter and over 350m in depth, the cenote Zacatón in central Mexico is a unique flooded sinkhole.A platform for conducting preliminary sonar tests is tethered in place. AbstractWe describe a Simultaneous Localization and Mapping (SLAM) method for a hovering underwater vehicle that will explore underwater caves and tunnels, a true three dimensional (3D) environment. Our method consists of a Rao-Blackwellized particle filter with a 3D evidence grid map representation. We describe a procedure for dynamically adjusting the number of particles to provide real-time performance. We also describe how we adjust the particle filter prediction step to accommodate sensor degradation or failure. We present an efficient octree data structure which makes it feasible to maintain the hundreds of maps needed by the particle filter to accurately model large environments. This octree structure can exploit spatial locality and temporal shared ancestry between particles to reduce the processing and storage requirements. To test our SLAM method, we utilize data collected with manually-deployed sonar mapping vehicles in the Wakulla Springs cave system in Florida and the Sistema Zacatón in Mexico, as well as data collected by the DEPTHX vehicle in the test tank at the Austin Applied Research Laboratory. We demonstrate our mapping and localization approach with these realworld datasets.

Abstract-The outdoor perception problem is a major challenge for driver-assistance and autonomous vehicle systems. While these systems can often employ active sensors such as sonar, radar, and lidar to perceive their surroundings, the state of standard traffic lights can only be perceived visually. By using a prior map, a perception system can anticipate and predict the locations of traffic lights and improve detection of the light state. The prior map also encodes the control semantics of the individual lights. This paper presents methods for automatically mapping the three dimensional positions of traffic lights and robustly detecting traffic light state onboard cars with cameras. We have used these methods to map more than four thousand traffic lights, and to perform onboard traffic light detection for thousands of drives through intersections.

Abstract-We present a method for infrastructure-free localization of underwater vehicles with multibeam sonar. After constructing a large scale (4 km), high resolution (1 m) bathymetric map of a region of the ocean floor, the vehicle can use the map to correct its gradual dead-reckoning error, or to re-localize itself after returning from the surface. This ability to re-localize is particularly important for deep-operating vehicles, which accumulate large amounts of error during the descent through the water column. We use a 3D evidence grid, stored in a efficient octree data structure, to fuse the multibeam range measurements and build maps that do not rely on particular features and are robust to noisy measurements. We use a particle filter to perform localization relative to this map. Both map and filter are general, robust techniques, and both run in real-time. Localization convergence and accuracy are improved, particularly over sparsely varying terrain, by deliberately selecting actions that are predicted to reduce the vehicle's position uncertainty. Our approach to this active localization is to select actions that are expected to generate sonar data that maximally discriminates between the current position hypotheses. Maximal discrimination is a very fast proxy for standard particle filter entropy-based active localization. These methods are demonstrated using a dataset provided by the Monterey Bay Aquarium Research Institute from their mapping AUV, collected near the Axial Seamount in the Juan de Fuca Ridge. Though it depends on the situation, the vehicle's position estimate typically converges to within 2 m of the true position in less than 100 s of traverse, or 150 m at 1.5 m/s. We explore the limitations of our approach, particularly with a smaller number of range sensors: although performance is degraded, satisfactory results are achieved with just four sonars.

This paper describes the application of a RaoBlackwellized Particle Filter to the problem of simultaneous localization and mapping onboard a hovering autonomous underwater vehicle. This vehicle, called DEPTHX, will be equipped with a large array of pencil-beam sonars for mapping, and will autonomously explore a system of flooded tunnels associated with the Zacatón sinkhole in Tamaulipas, Mexico. Due to the threedimensional nature of the tunnels, we describe an extension of traditional two dimensional evidence grids to three dimensions. In May 2005, we collected a sonar data set in Zacatón. We present successful SLAM results using both the real-world data and simulated data.

The deep phreatic thermal explorer (DEPTHX) is an autonomous underwater vehicle designed to navigate an unexplored environment, generate high-resolution three-dimensional (3-D) maps, collect biological samples based on an autonomous sampling decision, and return to its origin. In the spring of 2007, DEPTHX was deployed in Zacatón, a deep (approximately 318 m), limestone, phreatic sinkhole (cenote) in northeastern Mexico. As DEPTHX descended, it generated a 3-D map based on the processing of range data from 54 onboard sonars. The vehicle collected water column samples and wall biomat samples throughout the depth profile of the cenote. Post-expedition sample analysis via comparative analysis of 16S rRNA gene sequences revealed a wealth of microbial diversity. Traditional Sanger gene sequencing combined with a barcoded-amplicon pyrosequencing approach revealed novel, phylum-level lineages from the domains Bacteria and Archaea; in addition, several novel subphylum lineages were also identified. Overall, DEPTHX successfully navigated and mapped Zacatón, and collected biological samples based on an autonomous decision, which revealed novel microbial diversity in a previously unexplored environment.