Postdoc Robust Machine-Learning-Based Detection and Classification for Autonomous Underwater Robots

Updated: about 2 months ago
Deadline: 15 Feb 2024

In this project we will develop novel and robust machine-learning-based approaches for detection and classification for underwater robots using multi-sensor data from cameras, sonar, and magnetic sensors. The overall aim is to collectively detect and afterwards remove unwanted objects from coastal waters and seabeds. In this way the project contributes to keeping coastal waters clean, to protecting the environment, and to realizing global sustainability objectives.

This postdoc project is part of the Horizon Europe project SeaClear2.0 (Scalable full-cycle marine litter remediation in the Mediterranean: Robotic and participatory solutions, see also ).

The goal of SeaClear2.0 is to develop a collaborative multi-robot solution engaged in collecting marine waste using autonomous underwater robots. The aim is to realize efficacious marine litter detection and collection, while at the same time minimizing impact on underwater flora and fauna like seaweed and fish. This goal will be reached by bringing together state-of-the-art technologies from the fields of machine learning, control. optimization, and marine technologies and by building a stable and trustworthy system that is able of tackling sea and ocean pollution.

In this project we will focus on two major topics, where more emphasis can be put on either topic based on the expertise and interests of the selected candidate:

  • model-based and learning-based approaches for multi-sensor data fusion and detection and classification of underwater litter, and/or
  • robust detection and classification of multi-sensor underwater data in the presence of limited training data.
  • For topic (1) deep learning, physics-informed learning, and multi-sensor fusion will be the principal solution directions where the aim is to combine and merge information from cameras, sonar, magnetic sensors to detect and classify litter, fish, seaweed, etc. In addition, by integrating model-based and learning-based decision making approaches we will be able to use a priori information from dynamical or behavior models of fish, seaweed, plastic, etc. to significantly enhance detection and classification compared to existing approaches.

    For topic (2) the aim is to develop efficient methods to detect and classify underwater litter from cameras, sonar, magnetic sensor data that are robust to disturbances, glare effects, different light intensities, shadows, etc. Moreover, in view of the limited availability of labeled underwater images (and even less so for sonar or magnetic data) another important challenge is to do this with a limited amount of training data. In this context, transfer fusion learning where labeled image data are transformed into label data for sonar data of the same scene is another challenging research topic.

    The postdoc will join our machine learning and optimization team at the Delft Center for Systems and Control (DCSC) of Delft University of Technology. At the DCSC, our mission is to conduct foundational research in systems dynamics and control, involving dynamic modeling, advanced control theory, and optimization with societally important application fields including energy, transportation, and sustainability. 

    We offer the opportunity to do scientifically challenging research in a multi-disciplinary research group.

    This position is perfect for you if you have a PhD degree in systems and control, computer science, AI, applied mathematics, or a related field, and with a strong background in machine learning and decision making or control. You are also expected to work on the boundary of several research domains. A good command of the English language is required.

    Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities (salary indication: € 4.036 - € 5.090 per month gross). The TU Delft offers a customisable compensation package, discounts on health insurance, and a monthly work costs contribution. Flexible work schedules can be arranged.

    For international applicants, TU Delft has the Coming to Delft Service . This service provides information for new international employees to help you prepare the relocation and to settle in the Netherlands. The Coming to Delft Service offers a Dual Career Programme for partners and they organise events to expand your (social) network.

    The position is a temporary assignment for up to 24 months.

    Delft University of Technology is built on strong foundations. As creators of the world-famous Dutch waterworks and pioneers in biotech, TU Delft is a top international university combining science, engineering and design. It delivers world class results in education, research and innovation to address challenges in the areas of energy, climate, mobility, health and digital society. For generations, our engineers have proven to be entrepreneurial problem-solvers, both in business and in a social context.

    At TU Delft we embrace diversity as one of our core values  and we actively engage  to be a university where you feel at home and can flourish. We value different perspectives and qualities. We believe this makes our work more innovative, the TU Delft community more vibrant and the world more just. Together, we imagine, invent and create solutions using technology to have a positive impact on a global scale. That is why we invite you to apply. Your application will receive fair consideration.

    Challenge. Change. Impact!

    From chip to ship. From machine to human being. From idea to solution. Driven by a deep-rooted desire to understand our environment and discover its underlying mechanisms, research and education at the 3mE faculty focusses on fundamental understanding, design, production including application and product improvement, materials, processes and (mechanical) systems.

    3mE is a dynamic and innovative faculty with high-tech lab facilities and international reach. It’s a large faculty but also versatile, so we can often make unique connections by combining different disciplines. This is reflected in 3mE’s outstanding, state-of-the-art education, which trains students to become responsible and socially engaged engineers and scientists. We translate our knowledge and insights into solutions to societal issues, contributing to a sustainable society and to the development of prosperity and well-being. That is what unites us in pioneering research, inspiring education and (inter)national cooperation.

    Click here  to go to the website of the Faculty of Mechanical, Maritime and Materials Engineering. Do you want to experience working at our faculty? These videos  will introduce you to some of our researchers and their work.

    For more information about this vacancy, please contact Bart De Schutter, [email protected] .

    The position can either be a full-time one, or if the successful candidate requests it, a part-time one (80% or higher). In accordance with the equal opportunity policy of Delft University of Technology female candidates are in particular encouraged to apply.

    Are you interested in this vacancy? Please apply by 8 January 2024 via the application button and upload your letter of application along with a detailed curriculum vitae, a motivation why the proposed research topic interests you, a list of publications, (electronic) copies of your three most relevant journal or conference publications, the abstract and/or summary of your PhD thesis, your MSc course program and the corresponding marks, names and addresses of three reference persons, and all other information that might be relevant to your application. 

    For information about the application procedure, please contact Irina Bruckner, our HR advisor at [email protected] .

    Please note:
    - A pre-employment screening can be part of the selection procedure.
    - You can apply online. We will not process applications sent by email and/or post.
    - Acquisition in response to this vacancy is not appreciated.   

    Similar Positions