Supervisor: Sebastian Stein
Co-supervisor Enrico Gerding, Paul Brookes (Siemens)
Connected and autonomous vehicles have the potential to make better use of the existing road infrastructure. Furthermore, users have different preferences, such as travel time, congestion and price, as well as environmental factors such as pollution. The aim of this project is to take these factors into account, and optimise routing based on predicted congestion and personal preferences. In addition, incentive mechanisms will be created to incentivise users to follow suggested routes, which may be socially optimal but may not be in the individual’s best interest.
In particular, the project will integrate the following strands of research:
1. Congestion prediction. To accurately predict future congestion, the system will use routes and destinations shared by other users of the system. At the same time, the system needs to consider privacy and ensuring a privacy-by-design approach.
2. Preference learning. Asking users for their preferences directly can be distracting and they may be unable to express them properly. A better way is to show different options and ask users to choose, enabling learning of preferences. The project will do so using reinforcement learning.
3. Route optimisation. Based on the prediction and the personal preferences of individual users, the system will come up with route suggestions that are "socially optimal" (i.e. maximise the combined welfare of all drivers). This optimisation involves not only not only travel time, but other preference dimensions such as pollution levels and even availability of electric vehicle charging stations.
4. Micro payments for incentives. Game theory/mechanism design techniques will be used to create payments that incentivise users to adhere to the recommendation. This work builds on the mechanism design applications in related work, particularly on electric vehicle charging.
5. Implementation using the Blockchain based on micropayments. A proof of concept will be implemented to create a demonstrator using a crypto currency to manage the micropayments.
Depending on the interest of the applicant, the project can emphasize on some aspects from the above directions.
Industrial Collaboration with Siemens
This PhD project is in collaboration with Siemens who are a key player in connected and autonomous vehicles. This is a unique opportunity to work with a global company. During your PhD you will be mainly based at Southampton University, but are expected to spend at least 12 weeks at Siemens over the 4-year PhD period. The exact arrangements can be flexible depending on the direction of the research and the interests of the applicant.
This project combines aspects of artificial intelligence, including machine learning and multi-agent systems, with behavioural economics and game theory. Good programming skills are essential.
Good undergraduate degree (minimum UK 2:1 honours degree, or its international equivalent).
Closing date: 31 August 2020
Funding: full tuition plus, for UK students, an enhanced stipend of £18,500 tax-free per annum for up to 4 years.
How To Apply
Applications should be made online selecting “PhD Computer Science (Full time)” as the programme. Please enter Enrico Gerding under the Topic or Field of Research.
Applications should include:
Two reference letters
Degree Transcripts to date
Apply online: https://www.southampton.ac.uk/courses/how-to-apply/postgraduate-applications.page
For further information please contact: [email protected]