Abstract
The thesis considers the problem of scaling autonomous vehicles at a time when there are only autonomous vehicles in operation. If the current population of vehicles were all autonomous, there would be forty million of them in the UK and 1.5 billion cars globally. Currently, scaling is not being addressed coherently. The solutions offered by standardised scenarios are being undertaken by manufacturers who are fundamentally in competition with each other and are pursuing the goal of market dominance. Other bodies are addressing other elements of the problem.There are unrelated technology gaps which preclude development of a comprehensive solution, primarily, the lack of: a context, the purpose and goals of the RTS (RTS), consideration of the expectations and behaviours of other road users, the digital security of the RTS as part of the critical national infrastructure (CNI), and a definition of what autonomous vehicles actually contribute to the purpose and goals of the RTS. These aspects are addressed within the thesis and a solution is presented in Chapter 11 as an Axiom system, specification and functions.
The solution provides a context and structure that enables scaling, and a digital security solution which is transferable to other elements of the CNI. It highlights that interoperability of developed systems is critical in enabling vehicles to operate in differing jurisdictions with different societal and behavioural expectations of the RTS. It respects and provides for preservation of investment in legacy systems which currently exist in the RTS, such as the traffic management system. Ultimately it identifies the role of accountability and responsibility for the elements that comprise the RTS, and the role of the Government in the successful deployment of autonomous vehicles at scale.
The strength of the thesis is in the use of formal methods to analyse and form the basis of the solution. Simplification was achieved, and the closed, zero trust RTS for autonomous vehicles reduced to a complicated system. It recognises in any interaction between cyberphysical vehicles and other road users, the RTS becomes a complex system at the point of interaction which introduces issues of computational complexity. An answer might be to define unique RTS for cyberphysical and other road users, but due to the potentially stochastic nature of their behaviour, they were determined to be outside of scope of the thesis and a subject for further work.
| Date of Award | 2025 |
|---|---|
| Original language | English |
| Sponsors | KESS 2 PhD Student, University of South Wales |
| Supervisor | Richard Ward (Supervisor) & Eric Llewellyn (Supervisor) |