Improving radiological monitoring using drones

Geelen Stef


Schroeyers Wouter, (UHasselt),

SCK•CEN Mentor

Camps Johan
+32 14 33 27 61

SCK•CEN Co-mentor

Vidmar Tim
+32 14 33 21 10

Expert group

Crisis Management and Decision Support

PhD started


Short project description

Accidental or deliberate CBRN1 events are considered as low probability events that might however have a big impact on citizens and society. In this context, the early detection and identification of CBRN hazards in a standardized way is of high importance. First responders are introducing new technologies, from which the use of drones2, equipped with a number of sensors, is one of the very challenging topics to date. Drones have the potential to remotely investigate, in an efficient way and at locations difficult to reach by first responders, a large number of hazards at the scene. In this context mainly VTOL (Vertical Taking Off and Landing) drones are considered, operated by the first responders near the scene and allowing to hoover for more detailed investigations if required. However, the limited pay-load of these systems pose important challenges related to the required efficiency of the detection systems. The sensitivity of the sensors, operated in flight, should meet pre-defined operational detection levels, allowing to assess the exposure of first responders and citizens and to compare with certain health standards in real-time.

Within this Ph.D. topic the focus will be fully on RN sensors for alpha, beta and gamma detection and only direct measurements will be explored (no sampling). The characterization of the performance of detector systems coupled to drones and used in-flight poses several research challenges. Firstly, source – detector distances can be large and in addition sources can comprise large areas or volumes. This makes classical approaches such as standard Monte Carlo simulations very inefficient. Work has been done in this respect and has solved the problem for point sources and recently methods for surface sources are available. A method for a large surface source was integrated in the Penelope Monte Carlo code. However for other configurations, such as a radioactive cloud inside a building or in the outside environment, originating from a dispersal device, little work has been done to date. Secondly, it is not known to what effect rotor movement would influence detection performance, especially for direct alpha and beta monitoring, since the limited range of these types of radiation. Finally, a moving detector poses opportunities, but also limitations such as the short measurement time at a certain location.

The overall goal of the Ph.D. would be to contribute to improve the calculation tools for simulating the RN detector response within the typical (and potential future) operational range of VTOL drones. Based on this, conceptual detector designs can be proposed and/or evaluated. SCK•CEN started recently some experimental R&D work on drones, by coupling a small volume gamma-spectrometry detector to a drone, in a first phase mainly to map radioactive ground contaminations and/or locate a radioactive source. Experience from this work, which is expected to be gained in the next years, can be used within the Ph.D. to test and validate some of the numerical simulations methods.


1CBRN: Chemical, Biological, Radiological and Nuclear, sometimes expanded to include explosives: CBRNE

2We will use the term drone throughout this proposal, sometimes other terminology is preferred such as: UAV (Unmanned Aerial Vehicle) or RPAS (Remotely Piloted Aircraft System)


The Ph.D. work will start with the identification of a set of deliberate RN scenario’s ranging from radiological exposure devices to radiological dispersal devices. Based on these scenario’s and the potential hazards, the detection requirements will be determined. These detection requirements will be used as a reference for the evaluation of existing and proposed detection systems.

In parallel, a literature review will be made of existing detector response simulation methodologies for the different source detector geometries encountered in the RN scenarios identified and studied. In this context, especially the work done within the EURAMET-MetroERM project will be used as a starting point and an evaluation will be made what type of methods are necessary for improved simulation of drone detection capabilities in different scenario’s.

The main part of the PhD will concentrate on the development of improved methods for the simulation of source–detector geometries encountered in RN drone monitoring. These methods will include Monte Carlo techniques, but analytical techniques are not excluded. Detection systems and monitoring methodologies, including possible flight plans, will be designed for selected scenario’s to demonstrate the usefulness of the methods developed and/or as an iterative process in the research towards improved simulation methodologies.

In addition, finite element simulations, using COMSOL MultiPhysics or another similar package, will be undertaken to study the effect of rotor movement on radionuclide concentrations near possible detector locations on the drone, to estimate the effect on direct alpha and beta detections and complement the radiation transport simulations and methodologies.

Finally, based on the work performed, experimental datasets or new experiments will be identified to test and validate the methodologies developed. Experimental data-sets expected to be obtained in the next years from the drone programme started at SCK (first flights have been executed in October 2017) can be used.