Thesis
Multi-objective trajectory optimisation under epistemic uncertainty
- Creator
- Rights statement
- Awarding institution
- University of Strathclyde
- Date of award
- 2026
- Thesis identifier
- T17583
- Person Identifier (Local)
- 201989971
- Qualification Level
- Qualification Name
- Department, School or Faculty
- Abstract
- This thesis proposes innovative methodologies for the many-objective optimisation ofspacecraft trajectories under epistemic uncertainty, which can also have applicationoutside the domain of trajectory optimisation. Epistemic uncertainty refers to a typeof uncertainty that is due to a lack of knowledge, and which cannot be adequatelyquantified using standard probability theory, making it more challenging to quantifyin practice, as computational costs can become prohibitively high, in particular whenthe number of uncertain variables is increased. Thus, the main challenge addressed bythis thesis was reducing the computational cost of quantifying epistemic uncertainty inthe context of trajectory optimisation. Several strategies were developed and used incombination.Since the quantification of epistemic uncertainty is an integral part of the objectives being optimised, strategies are developed that reduce the number of evaluationsof these objectives. A surrogate model of the lower expectation is combined with adimensionality reduction technique to contain the computational cost. These dimensionality reduction techniques take the form of control maps, which map the space ofcontrol laws to a lower dimensional space, reducing the dimensionality of the searchspace. Several control maps are tested and compared.The second approach is to reduce the computational cost of quantifying the epistemic uncertainty. Techniques are proposed for some general formulations of epistemicuncertainty. These are accompanied with studies of the scalability of their computational cost with the number of epistemic variables.To optimise trajectories while accounting for the responsiveness of the control lawto observations of the state during the trajectory, sampling the uncertain variables and finding the optimal control law for each case can become prohibitively expensive, evenin the context of standard probability theory. Dynamic programming strategies havebeen applied, in the literature, to stochastic optimal control. A key innovation in thiswork is a dynamic programming formulation for epistemic uncertainty. This allowsdesigning trajectories where the control law responds to observations and corrects fordeviations, while still efficiently quantifying the epistemic uncertainty. Additionally,by splitting the trajectory into segments affected by different epistemically uncertainvariables, this limits the growth of the computational cost with the number of epistemicvariables. Additional key contributions were theoretical proofs that bound the value ofthis estimate with respect to other epistemic uncertainty quantifications of interest, in away which is applicable to many commonly used formulations of epistemic uncertainty.These methodologies are incorporated into algorithms which are applied to testcases consisting of spacecraft equipped with low thrust engines, whose parameters areaffected by epistemic uncertainty. These consist of an asteroid rendezvous mission, andan asteroid fly-by tour of four asteroids in the asteroid belt. For these, a set of controllaws are obtained, describing the evolution of the thrust magnitude and direction overtime, and launch parameters, which are robust with respect to epistemic uncertainty inthe system parameters. The algorithms used are empirically and theoretically shownto scale well with the number of uncertain variables.
- Advisor / supervisor
- Vasile, Massimiliano
- Resource Type
- DOI
- Funder
Relations
Items
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
PDF of thesis T17583 | 2026-02-10 | Public | Download |