angelus mallius

Angelos Mallios defends his PhD Thesis “Sonar Scan Matching for Simultaneous Localization and Mapping in Confined Underwater Environments”

This thesis presents the development of a localization and mapping algorithm for an autonomous underwater vehicle (AUV). It is based on probabilistic scan matching of raw sonar scans within a pose-based simultaneous localization and mapping (SLAM) framework.   To address the motion-induced distortions affecting the generation of full sector scans, an extended Kalman filter (EKF) is used to estimate the robot motion during that scan. The filter uses a constant velocity model with acceleration noise for motion prediction.Velocities from Doppler velocity log (DVL) and heading measurements from attitude and heading reference system (AHRS) are fed asynchronously and update the state….


Coverage Path Planning for Autonomous Underwater Vehicles

Doctoral thesis “Coverage Path Planning for Autonomous Underwater Vehicles” By Enric Galceran, PhD student of the Doctoral Program in Technology Supervised by Dr. Marc Carreras Pérez   Abstract At present, a mission to survey the ocean floor with an Autonomous Underwater Vehicle (AUV) is typically planned by selecting a list of waypoints that then the vehicle will automatically navigate through while keeping a safe distance from the bottom. Nonetheless, this approach has major drawbacks: (1) it does not allow the vehicle to safely operate amidst protrusions on the sea floor; (2) when traversing rugged terrain, the vehicle is forced to…


Automated underwater object classification using optical imagery

PhD Thesis “Automated underwater object classification using optical imagery”   By Shihavuddin, A.S.M Supervised by Dr. Nuno Grácias, Dr. Rafael García   Abstract This thesis addresses the problem of automated underwater optical image characterization. Remote underwater optical sensing allows the collection and storage of vast amounts of data for which manual classification may take months. Supervised automated classification of such datasets can save time and resources and can also enable extraction of valuable information related to marine and geological research.


Computer-aided lesion detection and segmentation on breast ultrasound

Doctoral thesis “Computer-aided lesion detection and segmentation on breast ultrasound”   By Gerard Pons Rodríguez. Supervised by Joan Martí Bonmatí, Robert Martí Marly.   Abstract This thesis deals with the detection, segmentation and classification of lesions on sonography. The contribution of the thesis is the development of a new Computer-Aided Diagnosis (CAD) framework capable of detecting, segmenting, and classifying breast abnormalities on sonography automatically. Firstly, an adaption of a generic object detection method, Deformable Part Models (DPM), to detect lesions in sonography is proposed. The method uses a machine learning technique to learn a model based on Histogram of Oriented…