Computer Assisted Decision Making in Combat Simulation
Location: Adelaide, SA (with remote options)
Duration: 4-6 months
Please note: Due to the sensitivity and security of this project, students must have Australian Citizenship to apply. Any applicants not meeting this requirement will automatically be deemed ineligible for this project.
DST Group models combat encounters using combat simulation to explore the potential impact and interactions of new or modified technologies, tools or tactics. Further, simulations represent realistic uses of tactics and force mixes to ensure useful insights can be gained from the outputs. The outcome of the simulation can be sensitive to tactics, positioning and routes, in addition to the given technologies or force structures being directly tested by a given experiment.
Currently, many aspects of the development process rely on hand crafted solutions refined by experimentation. This can be a time consuming process. Further, solutions that work for an isolated engagement may not balance allocation of resources, or positioning across multiple engagements. Similarly, ideal tactics will change dependant on overall mission goals for the larger scenario.
An ideal solution would be able to take in mission parameters, scenario information and recommend possible tactics that can be further refined or connected to subject matter expert solutions.
Research to be Conducted
The project will design an implementation of a machine learning solution that will provide tactical advice for selecting and coordinating ambush locations in a scenario. The project will aim to perform this given the following key deliverables.
- Deliverable 1: A software specification detailing a solution for identifying potential ambush locations, given environmental and platform data, on a given map. Locations should include coordination of resources within the ambush as well as key parameters for the engagement.
- The solution will describe the necessary software to be built or deployed, any data requirements and any additional resources for deployment of such a system
- The solution will be designed to work with our existing Java/ Python simulation frameworks
- The recommended solution will not need to be built within the internship, however the report should provide sufficient detail for a later implementation. This may include using appropriate level UML (Unified Modelling Language) diagrams to describe the software design.
- Deliverable 2: An extension outlining a planning system to recommend coordination of ambush locations with other mission goals.
- The planning system should be able to either take in the output of the proposed solution in deliverable 1, or a manually generated initial set of points and requirements
- The solution should consider how to scale the design to work with either high performance computing(HPC), or distributed computing resources
We are looking for a PhD student with the following:
- Expertise in software engineering, computer science or related software development discipline
- Familiarity with machine learning techniques
- Experience in geospatial information systems, computer vision, or related terrain analysis
- Experience in computer aided decision making
- Experience with machine learning libraries and frameworks
- Familiarity with Java and Python
- Familiarity with distributed computing and/or HPC, including GPU based techniques
Specific deliverables will be negotiated, dependant on the researcher’s skill set.
Essential outcomes include:
- A literature review of existing algorithms, software, concepts and principles that are applicable to the problem.
- An initial proposal outlining the proposed solution as well as detailing current gaps in resources or capability for implementation
- A detailed design document with sufficient detail for DST developers to implement the proposed solution
The intern will receive $3,000 per month of the internship, usually in the form of stipend payments.
It is expected that the intern will primarily undertake this research project during regular business hours, spending at least 80% of their time on-site with the industry partner. The intern will be expected to maintain contact with their academic mentor throughout the internship either through face-to-face or phone meetings as appropriate.
The intern and their academic mentor will have the opportunity to negotiate the project’s scope, milestones and timeline during the project planning stage.
30 October 2019
APR – 1116