Ferri
Gabriele
Ferri
Gabriele
No Thumbnail Available
Search Results
Now showing
1 - 2 of 2
-
PreprintMapping multiple gas/odor sources in an uncontrolled indoor environment using a Bayesian occupancy grid mapping based method( 2011-06) Ferri, Gabriele ; Jakuba, Michael V. ; Mondini, Alessio ; Mattoli, Virgilio ; Mazzolai, Barbara ; Yoerger, Dana R. ; Dario, PaoloIn this paper we address the problem of autonomously localizing multiple gas/odor sources in an indoor environment without a strong airflow. To do this, a robot iteratively creates an occupancy grid map. The produced map shows the probability each discrete cell contains a source. Our approach is based on a recent adaptation [15] to traditional Bayesian occupancy grid mapping for chemical source localization problems. The approach is less sensitive, in the considered scenario, to the choice of the algorithm parameters. We present experimental results with a robot in an indoor uncontrolled corridor in the presence of different ejecting sources proving the method is able to build reliable maps quickly (5.5 minutes in a 6 m x 2.1 m area) and in real time.
-
PreprintA novel trigger-based method for hydrothermal vents prospecting using an autonomous underwater robot( 2010-04) Ferri, Gabriele ; Jakuba, Michael V. ; Yoerger, Dana R.In this paper we address the problem of localizing active hydrothermal vents on the seafloor using an Autonomous Underwater Vehicle (AUV). The plumes emitted by hydrothermal vents are the result of thermal and chemical inputs from submarine hot spring systems into the overlying ocean. The Woods Hole Oceanographic Institution's Autonomous Benthic Explorer (ABE) AUV has successfully localized previously undiscovered hydrothermal vent fields in several recent vent prospecting expeditions. These expeditions utilized the AUV for a three-stage, nested survey strategy approach (German et al., 2008). Each stage consists of a survey flown at successively deeper depths through easier to detect but spatially more constrained vent fluids. Ideally this sequence of surveys culminates in photographic evidence of the vent fields themselves. In this work we introduce a new adaptive strategy for an AUV's movement during the first, highest-altitude survey: the AUV initially moves along pre-designed tracklines but certain conditions can trigger an adaptive movement that is likely to acquire additional high value data for vent localization. The trigger threshold is changed during the mission, adapting the method to the different survey profiles the robot may find. The proposed algorithm is vetted on data from previous ABE missions and measures of efficiency presented.