PhD position: Collective behaviour of autonomous organisms: from bio-particles to robotics

The project is supervised by Dr Fabien Paillusson (lead), Dr Alan Millard and Prof Andrei Zvelindovsky

Students who are interested in applying are advised to get in touch with Dr Fabien Paillusson <>.

Deadline for application is February 28, 2021 and more details about the application process can be found here

About the Project

This project aims to develop physical models of collective behaviour derived from biological systems comprising self-propelling agents, such as schools of fish and flocks of birds, and to then transform them into swarming behaviours of mobile robots. The ultimate aim is to improve the design of complex behaviours for robot swarms by integrating elements of collective intelligence inspired by biological agents.

Active Matter is an emerging interdisciplinary field in physics and applied mathematics that refers to systems comprising interacting agents that can drive their own motion (such as birds, fish, insects, “smart” artificial micro-particles, or bio-mimicking robots). Active Matter systems are opposed to Inert Matter systems, whose behaviours are entirely determined by the mechanical interactions between the agents. Consequently, in the past two decades Active Matter models have demonstrated complex collective behaviours such as the formation of active clusters, obstacle induced phase separation, and organised flocking motions, which are usually not achievable in assemblies of inert agents.

These newly found “living structures” can in turn be implemented in real life with collections of bacteria, artificial micro-particles, or bio-mimicking robots for industrial, medical or agricultural applications making use of their self-assembling properties and resilience to external influences. In Active Matter systems, simple sets of rules can lead to many rich phases of collective behaviours. There is ample opportunity to develop new classes of rules that can give rise to never-before-seen phases, and ultimately provide insights into how to reverse-engineer rules for targeted goals.

Active Matter systems are of particular interest with respect to swarm robotic systems where the behaviour of individual robots is affected by the fluid dynamics of their environment; for example, aerial drones whose position may drift when buffeted by gusts of wind, or surface-water / underwater vessels that may be dragged along by water currents. These swarm robotic systems must coordinate their movement despite the influence of the fluid environment they inhabit. By translating the mechanics of Active Matter systems to swarm robotic systems, we hope to improve the performance of embodied agents used in real-world applications. This novel approach would be particularly beneficial in GPS-denied environments such as deep-sea exploration, where individual robots must remain aggregated without the aid of an external frame of reference.

This interdisciplinary project at the interface of physics, computational modelling, and robotics will develop new theoretical and computational models for such systems and validate them on physical robotic swarms. It will broadly consist of three main tasks:

  • Formulating a set of microscopic rules deduced from a set of bio-inspired Active Matter systems, and formulating a statistical physics description of a large groups of such Active Matter systems
  • Deriving continuous field models starting from the microscopic rules defined in task (1) to describe the collective behaviour of Active Matter
  • Validating the developed models through their implementation on aerial / surface-water robot swarms – initially in the widely-used ARGoS multi-robot simulator, and then on physical hardware

This project combines complementary expertise from two schools: the School of Mathematics and Physics and the School of Computer Science, building on existing research. The successful student will be associated with the Centre for Computational Physics and the Lincoln Centre for Autonomous Systems (L-CAS), of which the supervisory team are members.

Skills the candidate will learn:

  • Theoretical and computational modelling
  • Design of control algorithms for robot swarms
  • Interdisciplinary collaboration skills
  • National and international collaboration skills
  • Technical oral presentation and written communication skills

Ideal candidates:

Interested applicants should hold, at a minimum, a 2.1 degree in either Physics, Mathematics, Computer Science, Engineering, or a related discipline. Applicants with a relevant Master’s degree are particularly welcome. The candidate is expected to have good communication and teamwork skills, and must be motivated to learn new things. Interested applicants are encouraged to demonstrate any skills and/or experience relevant to the project subject area(s) of interest.

Who is eligible for funding?

Please make sure to check the eligibility criteria before you apply. Normally, a student must have no restrictions on how long they can stay in the UK and have been ordinarily resident in the UK for at least 3 years prior to the start of the studentship.  UK students will be eligible for a full studentship, covering the costs of Home fees, and a stipend to support living costs for 3.5 years. 

Although most DTP students must be UK residents, we also have an opportunity for an international (EU and non-EU) student. The international studentship award will be subject to eligibility, and also the availability of complementary funding (to provide the differential to the international fee rate). You should get in touch with the lead supervisor before applying this award.  


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.