Evaluating the impact of sewage discharges on the marine environment with a lightweight AUV.

Environmental studies that use small Autonomous Underwater Vehicles (AUVs) can survey wider and deeper areas, compared to traditional methods, at a reasonable cost. Thanks to the precise vehicle navigation systems, the data collected can be accurately geolocalized. Besides, lightweight vehicles can be deployed from the shore or from small boats and programmed by means of user-friendly graphical interfaces, thus reducing and simplifying the need of human resources and infrastructures. Based on such a technology, this paper presents a framework to assess the environmental impact of a marine sewage outfall set in the Bay of Palma (Mallorca, Spain). We report the results of the analysis of the images recorded in the course of six missions conducted with an AUV. The plan was designed after a microbiological analysis detected the presence of cyanobacteria in a sample of sand and water collected by scuba divers close to the sewer pipe mouth.

Inertial sensor self-calibration in a visually-aided navigation approach for a micro-AUV.

This paper presents a new solution for underwater observation, image recording, mapping and 3D reconstruction in shallow waters. The platform, designed as a research and testing tool, is based on a small underwater robot equipped with a MEMS-based IMU, two stereo cameras and a pressure sensor. The data given by the sensors are fused, adjusted and corrected in a multiplicative error state Kalman filter (MESKF), which returns a single vector with the pose and twist of the vehicle and the biases of the inertial sensors (the accelerometer and the gyroscope). The inclusion of these biases in the state vector permits their self-calibration and stabilization, improving the estimates of the robot orientation. Experiments in controlled underwater scenarios and in the sea have demonstrated a satisfactory performance and the capacity of the vehicle to operate in real environments and in real time.

Trajectory-based visual localization in underwater surveying missions.

We present a new vision-based localization system applied to an autonomous underwater vehicle (AUV) with limited sensing and computation capabilities. The traditional EKF-SLAM approaches are usually expensive in terms of execution time; the approach presented in this paper strengthens this method by adopting a trajectory-based schema that reduces the computational requirements. The pose of the vehicle is estimated using an extended Kalman filter (EKF), which predicts the vehicle motion by means of a visual odometer and corrects these predictions using the data associations (loop closures) between the current frame and the previous ones. One of the most important steps in this procedure is the image registration method, as it reinforces the data association and, thus, makes it possible to close loops reliably. Since the use of standard EKFs entail linearization errors that can distort the vehicle pose estimations, the approach has also been tested using an iterated Kalman filter (IEKF). Experiments have been conducted using a real underwater vehicle in controlled scenarios and in shallow sea waters, showing an excellent performance with very small errors, both in the vehicle pose and in the overall trajectory estimates.