RFFMIDT: A UAV SAR System for Oil Spill Detection in the Arctic (2014-2015)

Project partners

NTNU i Ålesund, Mechtronics Lab

project period

2014-2015

Project type

RFFMIDT

Short project description

Oil spills and accidental releases of oil into the ocean are of urgent concern because of damage to the marine and coastal ecosystems as well as the adverse effects that spills and releases have on humans, given that the human population depends on the sea. Nowadays, the likelihood of oil spills and accidental releases is increasing in the Arctic area as new shipping routes become available and oil resources become more easily accessible due to a reduction in the extent and thickness of the Arctic sea ice. Spill detection is especially important in the Arctic area as oil may be hidden under snow and ice when it may be almost completely dark the entire day. In order to address the increased risk, new knowledge and technology for oil spill detection in Arctic conditions are urgently needed. In recent years, a new remote sensing technology based on Unmanned Aerial Vehicle (UAV) SAR has emerged. This technology provides a great potential for detailed monitoring and surveillance of areas covering of up to a few thousand square kilometres. Relative to a manned aircraft, the advantages of using a UAV are that it does not need a qualified pilot on board, can enter environments that are dangerous to human life, can be programmed to work autonomously, performs a precise scan of a region during day and night and it can be built at a relatively low cost.

This project combines the research experience in radar signal processing and radar systems to do a theoretical and experimental study of the polarimetric SAR application in oil spills with ice, present novel decomposition and classification algorithms in full polarimetric SAR for the UAV platform, and design the scheme of the UAV SAR system.

sar

Objectives / aims for the project

This project lasts 12 months, we have the following secondary goals:

  1. Explore and verify multi-frequency and full-polarimtric SAR for oil-in-ice technology:
    • Microwave scattering models for different oil-ice interfaces with multi-frequency and full-polarimetirc mode
    • Full-polarimetric SAR decomposition and classification methods for different oil-ice interfaces
    • Experiment verification of the new models and methods in the oil-in-ice environment
  2. Analyse and design a UAV SAR system:
    • An optimised multi-frequency and full-polarimtric SAR system scheme for the UAV platform
    • Real-time UAV SAR processor design
  3. Establish a solid research group on underwater robotics at NTNU i Ålesund
    • Regular internal seminars on the topic and a workshop with international researchers
    • Pre-project management and report, and prepare and establish a 3-year research activity

Research content

  1. Arctic oil spills could result in different conditions: oil floating on the surface of the ocean, released into icy water, lying on or under the ice, entering into pockets or pools on the ice surface. In order to detect an oil spill in ice, it is necessary to identify some features of the oil that are more or less unique in a SAR image. Studying should begin by analysing the structure of oily ice, which would lead to an accurate model based on simulations. Then, the polarimetric scattering model at various microwave frequencies will be determined and constructed.Most operational oil detection methods are still based on the image intensity differences between the oil spill and surroundings using a single polarisation channel. As described above, full polarimetric SAR imagery has the potential for detecting oil because the reflections or scattering from ice-oil and ice-brine interfaces are different in the complex domain. It is necessary to determine whether available decomposition methods are appropriate or if new decomposition approaches should be developed according to the electromagnetic properties of the new models for different oil-in-ice conditions.Experimental verification of the new models and algorithms can first be carried out using data provided by current polarimetric SAR systems to detecting oil or ice individually. However, total verification of the model and algorithm using the existing SAR systems is difficult because the combined ice and oil data is rare and the relatively high NESZ of current polarimetric satellite SARs make such systems unideal for verifying the full polarimetric SAR models and algorithms in low S/N conditions. As such, an experiment site with ice and oil in different conditions should be set up. The ground or shipborne SAR system using instruments needs to be constructed first to test and gather data for full polarimetric conditions.
  2. In order to detect oil in the ice, a relatively low radar frequency in the L or P band would be advantageous to reduce attenuation of the exact frequency, which will be determined according to the models analysis and simulation. The radar transmitter and receiver structure will be analysed and determined as well as the antenna scheme and parameters, which will be studied to allow for the full-polarimetric function. In order to build a SAR system with low NESZ, the radar parameters will be analysed thoroughly. Besides new approaches to improve noise performance using direct conversion as well as calibration methods will be studied. High resolution motion compensation for relatively small UAV platforms is especially important. As such, the motion compensation scheme, including inertial measurements and GPS, will be designed and new SAR imaging algorithms and autofocus compensation methods suitable for UAV platforms will be studied.

Degree of innovation

This research project focuses on developing a multi-frequency and full-polarimetric UAV SAR, to make it adaptable to oil spill detection in ice. To the applicant’s knowledge, at the time of writing, there has been little work done on this topic. The project is devoted to solving the difficulties encountered with different oil-in-ice conditions known to Arctic environments. The project will promote both theoretical and practical solutions for oil spill detection and the technology used in this system will increase the possibilities of development in the UAV SAR area of research in Norway. The proposed research will pave the way to progress of the UAV SAR in other remote sensing areas as well.

Project impact

To date, as far as we are aware, there is no UAV SAR system available in Norway that can be used for effectively detecting oil spills in the Arctic environment. The proposed UAV SAR system will not only used in the oil spill application, but also allow for agricultural monitoring, disaster monitoring, mapping of aquaculture, forest, wetlands and so on, thereby improving the technological level of environmental monitoring and ecological protection. Meanwhile, since a research group will be founded at NTNU i Ålesund, we will train scientists (Ph.D, and Master) to be at the frontier of this international and highly competitive field. Workshops and seminars will be organised during the project. Also, at least two papers will be published for first-class related conferences or in journals on remote sensing or radar signal processing.

 

Comments are closed