PhD in Computer Vision and Machine Learning - DeeperSense H2020

Updated: about 2 months ago
Job Type: FullTime
Deadline: 15 Apr 2021

This position is offered under the framework of EU H2020 project DeeperSense. This project seeks a novel approach of using Artificial Intelligence and data-driven Deep Learning to combine non-visual and visual sensors with the objective of improving environment perception beyond the abilities of the individual sensors. We focus on underwater robotics as a domain to demonstrate and verify this approach, as it is considered one of the most challenging application areas for robot operation and environment perception.

A central goal of this PhD work  is to use advanced machine learning techniques to enable multisensor fusion of different sources, mainly from side scan sonars and multibeam echosounders. Optical imagery will be used as another layer for training and benchmarking with the overall objective of allowing accurate sea-floor characterization from an AUV without the need for optical imagery. Other topics of research will be tailored to the student’s interests and may include methods based on machine learning for processing forward looking sonar in mapping in 2D and 3D, explainable AI applied to sensors and others.

Tasks and Responsibilities

  • Development of novel perception methods for underwater robotics based on optics and acoustics.
  • Development of novel approaches for intersensoric learning.
  • Authoring and co-authoring of journal and conference articles
  • Communicating with our industrial partner (Tecnoambiente, SA) and participate on field experiments using the underwater robots of the University of Girona and those of the Deepersense consortium
  • Participation in project activities such as meetings and joint field experiments

Submission instructions and further information: https://vicorob.udg.edu/careers/ 


View or Apply

Similar Positions