Using Machine Learning to Test Robotic Systems

Project Description:

Software testing is the process of executing the system under test and checking that the observed outputs are acceptable. In practice, a systems tester will select test cases and examine their outcomes according to their own mental model of the system under test. However, this becomes problematic when the system in question is complex, with lots of components that might give rise to interactions that are hard to anticipate. With robotic systems the problem is exacerbated because an “output” amounts to a physical behaviour sequence of a machine, which can be difficult to capture and reason about.

In this project you will develop a new solution to this testing predicament. Instead of relying upon a human’s mental model of a system, this new solution will propose an automated testing approach, driven by Machine Learning. You will develop an approach that automatically and strategically tests a system in an incremental manner, building up an internal (learnt) `mental model’ of system behaviour in the process, so that the human tester can focus on determining whether the behaviour is correct or not.

There will be several novel aspects of this research, including the machine learning techniques used, the way in which test cases are generated from a learnt model, and the evaluation. Evaluation will be based on a mixture of experiments and case studies drawn from robotics. There is potential to collaborate with members of the RoboTest project and also become involved in Sheffield Robotics.


Application deadline: 30th August 2019

Supervisor: Professor Robert Hierons
Second Supervisor: Dr Neil Walkinshaw

Required Qualifications:

- Applicants should have at least a 2:1 in a BSc (hons) degree, in a related discipline, such as Computer Science.

- If English is not your first language, you must have an IELTS score of 6.5 overall, with no less than 6.0 in each component

How to apply:

To apply for the studentship, applicants need to apply directly to the University of
Sheffield using the online application system. Please name Professor Robert Hierons as your
proposed supervisor.

Complete an application for admission to the standard Computer Science PhD
programme
https://www.sheffield.ac.uk/postgraduate/phd/apply/applying

Applications should include a research proposal, CV, transcripts and two references.
The research proposal (up to 4 A4 pages, including references) should outline your
reasons for applying for this studentship and how you would approach the
research, including details of your skills and experience

Funding notes:

The studentship will commence on 30th September 2019. The studentship will cover tuition fees at the UK/EU rate and provide an annual stipend at the standard RCUK rate (currently £15,009 for 2019/20) for three and a half years. If you are an international student, you are eligible to apply but you must have the means to pay the difference between the UK/EU and Overseas tuition fees.

Informal enquiries should be directed to: Professor Robert Hierons ([Email Address Removed])

The Department of Computer Science, within the Faculty of Engineering, was established in 1982 and has since attained an international reputation for its research and teaching. In REF2014, the Department was ranked 5th out of 89 computer science departments in the UK, indicating that the large majority of our research is nationally or internationally leading. Further information about the Department can be found at http://www.sheffield.ac.uk/dcs.