This paper presents a method for processing sparse, non-Gaussian multimodal data in a simultaneous localization and mapping (SLAM) framework using factor graphs. Our approach demonstrates the feasibility of using a sum-product inference strategy to recover functional belief marginals from highly non-Gaussian situations, relaxing the prolific unimodal Gaussian assumption. The method is more focused than conventional multi-hypothesis approaches, but still captures dominant modes via multi-modality. The proposed algorithm exists in a trade space that spans the anticipated uncertainty of measurement data, task-specific performance, sensor quality, and computational cost. This work leverages several major algorithm design constructs, including clique recycling, to put an upper bound on the allowable computational expense - a major challenge in non-parametric methods. To better demonstrate robustness, experimental results show the feasibility of the method on at least two of four major sources of non-Gaussian behavior: i) the first introduces a canonical range-only problem which is always underdetermined although composed exclusively from Gaussian measurements; ii) a real-world AUV dataset, demonstrating how ambiguous acoustic correlator measurements are directly incorporated into a non-Gaussian SLAM solution, while using dead reckon tethering to overcome short term computational requirements.

Autonomous underwater vehicles (AUVs) pose significant communication challenges: vehicles are submerged for periods of time in which speed-of-light communication is impossible. This is a particular problem on low-cost AUV platforms, on which acoustic modems are not available to get vehicle state or provide re-deploy commands. We investigate one possible method of providing operators with a communication line to these vehicles by using noise underwater to both classify behavior of submerged vehicles and to command them. In this scheme, processing of data from hydrophone arrays provide operators with AUV mode estimates and AUVs with surface vehicle behavior updates. Simulation studies were used to characterize trajectories for simple transect versus loiter behaviors based on the bearing and time to intercept (TTI). A classifier based on K-nearest-neighbor with dynamic time warping as a distance metric was used to classify simulation data. The simulation-based classifier was then applied to classify bearing tracking data from passive tracking of a loitering AUV and bearing and TTI data from passive tracking of a transecting boat based on field array data. Experiment data was classified with 76 % accuracy using bearing-only data, 96% accuracy for TTI -only data and 99 % accuracy for combined classification. The techniques developed here could be used for AUV cuing by surface vessels and monitoring of AUV behavior.

Accurate localization is critical for a robotic vehicle to navigate autonomously. Conventional autonomous underwater vehicles (AUV s) typically rely on an inertial navigation system (INS) aided by a Doppler velocity log (DVL) in order to reduce the rate of positional error growth of dead-reckoning to a level suitable for reliable navigation underwater. The size, cost, and power requirements of these systems result in vehicles that are prohibitively large and expensive for multi-AUV operations. In this work we present the first results of closed-loop experiments using a miniature, low-cost SandShark AUV and a custom-designed, inexpensive acoustic system first described in our previous work. Results are validated using an independent LBL system, and indicate that our approach is suitably accurate to enable the self-localization of such AUVs without the use of an expensive DVL-aided INS. Self-localization is performed by obtaining acoustic range and angle measurements from the AUV to a single acoustic beacon using a vehicle-mounted passive hydrophone receiver-array, and fusing these measurements using a particle filter. A critical aspect of our approach that allows for real-time, closed-loop operation is the close coupling of conventional phased-array beamforming and particle filtering - this implementation detail reduces the computational complexity associated with our previously described two-stage beamforming plus particle filtering process, and consequently also enables an increase in particle count and an improvement in navigational accuracy. Experimental results are provided for two cases: first, absolute navigation in the case where the beacon is fixed at a known position; and second, relative navigation with a moving beacon, a novel operating paradigm for AUVs which promises to enable multi-AUV operations while maintaining bounded navigation error.

This paper presents an acoustic localization system for small and low-cost autonomous underwater vehicles (AUVs). Accurate and robust localization for low-cost AUVs would lower the barrier toward multi-AUV research in river and ocean environments. However, these AUVs introduce size, power, and cost constraints that prevent the use of conventional AUV sensors and acoustic positioning systems, adding great difficulty to the problem of underwater localization. Our system uses a single acoustic transmitter placed at a reference point and is acoustically passive on the AUV, reducing cost and power use, and enabling multi-AUV localization. The AUV has an ultra-short baseline (USBL) receiver array that uses one-way travel-time (OWTT) and phased-array beamforming to calculate range, azimuth, and inclination to the transmitter, providing an instantaneous estimate of the vehicle location. This estimate is fed to a particle filter and graph-based smoothing algorithm to generate a consistent AUV trajectory. We describe the complete processing pipeline of our system, and present results based on experiments using a low-cost AUV. To the authors' knowledge, this work constitutes the first practical demonstration of the feasibility of OWTT inverted USBL navigation for AUVs.

Characterization of seabed targets using networks of Autonomous Underwater Vehicles (AUVs) is of great interest for harbor security. The imaging sensors generally used for target localization and classification are expensive for outfitting multiple AUVs and produce data that is difficult to use for real-time onboard classification. An approach to this problem has been developed in which an acoustic source insonifies targets and AUVs with inexpensive hydrophone nose arrays are used to sample the resulting scattered fields. These scattered fields consist of distinctive radiation patterns, where the amplitude sampled by a receiving AUV for a particular target is determined by source frequency, aspect angle between source and target, and bistatic angle between source and receiver. AUV behaviors were developed to control the bistatic angles relative to a localized target between source and receiver vehicles and aspect angle between a target and the source. Two configurations were explored: bistatic, where the source is fixed and receiver moving, and multistatic, where both source and receiver are located on mobile autonomous vehicles. Design considerations for behaviors to sample different combinations of bistatic angle and aspect angle included keeping receiving arrays broadside to a target for uniform-quality data collection and obeying vehicle dynamics constraints. Behaviors for the bistatic configuration were demonstrated in AUV experiments, and behaviors for the multistatic case were demonstrated in simulation, including collaboration between source and receiver vehicles for controlling bistatic versus aspect angle.