Project description
Effective sensor placement is a fundamental problem in surveillance, communications, robotics and environmental monitoring. Classical approaches are often based on graph-theoretic domination and covering models, but these models typically assume idealised sensing conditions. Modern approaches attempt to account for obstacles such as buildings, terrain or other sources of occlusion. This project will investigate optimisation methods for sensor placement in environments with realistic visibility constraints. The research will extend classical domination and covering problems to obstacle-aware sensing models and will explore the impact of these constraints. Possible directions, depending on student interests, could include:- Development of exact, heuristic and metaheuristic optimisation algorithms.
- High-performance implementations with consideration of parallel and GPU computing.
- Analysis of solution quality, robustness and scalability.
- Matching to practical hardware capabilities and deployment constraints.
Assumed knowledge
Some coding experience required, preferably in MATLAB, Python, or C. The theoretical aspects of this problem include mathematical formulations and techniques that mainly come from linear algebra and statistics.
Note: You need to register interest in projects from different supervisors (not a number of projects with the one supervisor).
You must also contact each supervisor directly to discuss both the project details and your suitability to undertake the project.