PhD Studentship: Swarm exploration: multi-robot visual navigation through insect-inspired strategies (2018)

What you get

£14,777 tax free stipend plus UK/EU fees waiver.

Type of award

PhD

PhD project

Navigation is a vital task for autonomous robots, whether searching a disaster zone or exploring a planet.

Bees are champion navigators, travelling miles for food efficiently and accurately:

• what if we could give robots the navigation abilities of a bee?
• and what if we could share the learnt knowledge in multi-robot teams?

For instance, flying robots could rapidly survey an area, identify a person in difficulty and pass their knowledge to bigger, but slower, wheeled robots that can offer assistance.

Building on our insect-inspired work at Sussex [1-3] you will develop bio-inspired algorithms that allow such navigation in heterogeneous robot teams, opening up a new area of robot autonomy.

This exciting interdisciplinary project will form part of the EPSRC-funded Brains on Board program grant, a multi-university project in which we aim to create robots with the learning abilities of bees.

You will join the Sussex Brains on Board team and contribute to the overall vision of the project, while benefitting from:

• years of experience in bio-inspired algorithms
• cutting-edge facilities and equipment
• funding for conferences
• the team's multi-disciplinary expertise.

Detail

Truly autonomous robots must navigate long distances with low power consumption in industrial contexts where GPS is unavailable or unreliable (rural agriculture with limited hardware, space exploration, search and rescue).

Despite limited computational power and low-resolution eyes, insects can navigate long and complex routes using vision with a performance far beyond current robots'.

Inspired by this impressive performance, we have recently developed visual navigation methods which allow robots to navigate complex environments using efficient, low power mechanisms (e.g. [1-3]).

In this PhD project, you will develop new algorithms for heterogeneous robot teams, allowing them to efficiently explore and visually navigate large areas by sharing knowledge, despite differing perspectives, and thus take advantage of their different capabilities.

Leveraging the power of swarms of heterogeneous robot teams is a vital component of future robotics development.

However, how information can be usefully shared and combined between robots remains an open question. A key feature of our algorithms is that the information needed to travel a route is encoded economically within a neural network [2].

While this does not prohibit the transfer of route knowledge between robots, it is not obvious how to do it, in particular when the views of the robots are different. Animals are also limited in their ability to communicate this information.

In this project, you will therefore go beyond animals' capabilities by developing perspective independent route encodings, route communication, and collective exploration and navigation algorithms.

There are many possible directions you can take the project with the team, working with ground-based and/or flying robots, or even aspects of human guidance via wearable sensors.

To excel in this project, you will either have a:

• strong quantitative background with interest in bio-inspired solutions
• a biological background with strong maths and computing skills and a keen desire to improve them.

• References

  1. Graham, P. and Philippides, A. (2017). Vision for navigation: what can we learn from ants? Arthropod Structure & Development, 46(5), 718-722.1.

  2. Philippides, A., Graham, P., Baddeley, B. and Husbands, P. (2015). Using neural networks to understand the information that guides behavior: a case study in visual navigation. Artificial Neural Networks, 227-244. Chicago.

  3. Baddeley B, Graham P, Philippides A and Husbands P (2012) A Model of Ant Route Navigation Driven by Scene Familiarity. PLoS Comput Biol 8(1): e1002336

This project is one of a number ear-marked for funding by the University of Sussex Engineering and Physical Sciences Research Council (EPSRC) Doctoral Training Partnership to commence in September 2018.

This project is in direct competition with others for funding; the projects which receive the best applicants will be awarded the funding.

Eligibility

To be eligible, you need to:

• have an excellent academic record and should have received or be expected to receive a relevant first or upper-second class honours degree
• be a UK or EU student who has been ordinarily resident in the UK for the previous three years. If you are an EU student and you don't meet this criteria, you will only be eligible for the fee waiver.

If you are an overseas (non-EU) student, you are not eligible to apply for EPSRC funding, but you can apply if you have access to other sources of funding.

Deadline

30 June 2018 17:00

How to apply

Apply for a PhD in Informatics through the University of Sussex postgraduate application system.

When you apply, you should include a brief statement of your scientific interests and skills/experience for the mandatory "research proposal" section, including how they relate to this project (maximum two pages).

You should also indicate Professor Thomas Nowotny as your preferred advisor and clearly state the title of the studentship in the finance section.

Contact us

Email Professor Thomas Nowotny at T.Nowotny@sussex.ac.uk

Timetable

The deadline is 02/03/2018.