Research Fellow in Machine Learning for Engine Control

Department of Electrical and Electronic Engineering
School of Electrical, Mechanical and Infrastructure Engineering
University of Melbourne

Salary: PhD entry Level A.6 $87,415 -$93,830 p.a. plus 9.5% superannuation

The research fellow will join a team of academic staff and postgraduate students working on problems related to the control and calibration of diesel engines. The team maintains a longstanding partnership with Toyota Motor Company in this area.

This position is available for 2 years and will be reviewed at the end of this period.

The aim of the research is to explore and develop novel algorithms that are implementable in real-time to improve the fuel economy and emissions of diesel engines. Given the constraints present in the system, much of the work to date has focused on the development of model predictive control designs. One of the challenges in implementation of these controllers is the calibration to achieve the appropriate level of performance under different legislated limits and consumer demands for different markets. We intend to develop methods (based on a combination of machine learning and traditional optimisation methods) that can partially automate the calibration process for these advanced controller designs, leading ultimately to faster calibration of better performing engine controllers.

The research fellow must have a background in engineering or applied mathematics, with demonstrated expertise in modelling and control of dynamical systems and numerical optimisation. Experience with at least one of model predictive control; engine modelling and control; machine learning; and implementation of controllers on real systems is essential – while some background in more than one of these areas is highly desirable.

Candidates are encouraged to apply via this link.

Application Closing Date: 31 January 2018.
Start Date: As soon as possible from March 2018.

Research Fellow in Optimisation and Control

Department of Electrical and Electronic Engineering
School of Electrical, Mechanical and Infrastructure Engineering
University of Melbourne

Salary: PhD entry Level A.6 $87,415 -$93,830 p.a.) plus 9.5% superannuation

The research fellow will join a team of academic staff and postgraduate students working on problems pertaining to real-time decision making in dynamic systems. The team maintains a longstanding partnership with Defence Science and Technology Group in this area.

In this research, the investigation will focus on the coordination of a group of autonomous vehicles to achieve a common objective under uncertain and time varying operational considerations. The flexibility to be able to handle different scenarios and associated mission constraints will be an integral aspect of the research.

The aim of the research is to explore and develop novel algorithms that are implementable in real-time, can predict over a long horizon and make optimal decisions under given performance metrics while satisfying hard constraints. This will involve a combination of predictive control and machine learning (data-driven) techniques to optimally respond to rapidly changing conditions, and to disrupt the planning algorithms of adversarial actions.

The research fellow will have an outstanding background in Engineering or Applied Mathematics (or equivalent), and experience with the implementation of numerical methods and engineering applications of optimisation techniques in real-time control of dynamical systems with exposure to mathematical foundations of learning, graph theory, and combinatorial optimisation.

The research fellow will be located in the Department of Electrical and Electronic Engineering within the Melbourne School of Engineering, and collaborate with researchers and engineers internally and externally. In addition to preparing technical reports, research publications, and computer simulations, the research fellow may also have the opportunity to undertake teaching and student supervision is areas directly related to their research.

Candidates are encouraged to apply via this link.

Application Closing Date: 31 December 2017.
Start Date: As soon as possible from March 2018.

Open PhD Positions

Applications are invited for two PhD scholarships in the areas of (1) real-time decision making for autonomous systems in uncertain environments, (2) secure networked control systems.

The work will be based within the Control and Signal Processing (CSP) Lab, MIDAS Lab, the Department of Electrical and Electronic Engineering, the University of Melbourne. The student will be supervised by Dr Iman Shames (see https://imanshames.blog for more information) and co-supervised by another MIDAS Lab faculty member.

The ideal candidate should have a degree in electrical and electronic engineering, computer science, applied mathematics, or mechanical engineering with solid background in applied mathematics and particularly in numerical methods, control theory, and/or optimisation.

The PhD scholarship provides 30K AUD a year tax-free stipend for up-to four years (subject to passing the PhD candidature confirmation after 12 months) as well as tuition fees for the duration of studies. Additionally, up to 15K AUD will be made available for travel funding to visit other research institutes and attending international conferences.

MIDAS is a large and complementary group of academic researchers and postgraduates collaborating across multiple departments within the Melbourne School of Engineering in areas including automation, control, analytics, machine learning, and optimisation.

Expressions of interest are invited from candidates with (or who expect to gain) a first-class honours degree or an equivalent degree in engineering, computer science, physics, mathematics, or a related discipline. Interested candidates should contact Dr. Shames (ishames@unimelb.edu.au) directly and include (1) a brief statement of interest in this position with clear indication of preference for the 2 areas mentioned above (maximum 2 pages), (2) their detailed CV, (3) bachelors and masters/honours transcripts, and (4) names and contact details of at least two referees.

Application Closing Date: 1 December 2017.
Start Date: As soon as possible from March 2018.