FEIT Research Project Database

Computational mass transport

Project Leader: Peter Dower
Collaborators: Prof William McEneaney (UCSD)
Primary Contact: Peter Dower (pdower@unimelb.edu.au)
Keywords: advanced control architectures; applied control theory; optimisation; physical modelling; systems theory
Disciplines: Electrical & Electronic Engineering
Research Centre: Melbourne Systems Laboratory (MSL)

The action principle is a fundamental postulate that underlies conservation laws in physics. In conjunction with tools from optimal control, optimisation, convex analysis, and functional analysis, the action principle can be used to formulate and solve computationally complex trajectory planning problems, such as arising in spacecraft navigation problems involving orbital injection and rendezvous, where these conservation laws apply.

In this project, this action principle approach will be applied to trajectory planning problems that involve the motion of distributions of matter, such as gas plumes and particle clouds. Computational methods for these ‘mass transport’ problems will be developed, implemented, and demonstrated, using numerical methods arising in optimal control, optimisation, and the study of partial differential equations.