PhD studentship on Cognitive models of human traffic interaction for real-time behaviour interpretation - U of Leeds

[Matteo Leonetti]

[apologies for multiple postings]

Funded PhD position available at the university of Leeds.

Description: 

There is a strong push towards the development of driverless, automated vehicles (AVs). However, to enable full self-driving in complex, urban environments, AVs will need to participate in the subtleties of on-road interactions, appropriately interpreting and responding to the goals and intentions of human road users while at the same time communicating and pursuing the AVs’ own goals. 

In the Human Factors & Safety research group at the Institute for Transport Studies, University of Leeds, we are actively addressing this open research challenge in a number of ways, including the development of cognitively plausible mathematical models which quantitatively describe human interactive behaviours in traffic, and application of these models to understand and improve human-AV interactions. This PhD studentship, sponsored by Nissan Motor Manufacturing (UK) Limited, will allow the successful candidate to build further on the cutting edge models from our research group and elsewhere, and to connect it to state of the art methods for real-time AV perception and decision-making. The overarching goal is to implement models and algorithms that can estimate, from processed AV sensor data, what a given human road user perceives the AV’s near-term intentions to be. 

A preliminary list of intended intermediate objectives and activities includes: - Identifying interaction scenarios where an interaction model are likely to be most beneficial to real-time AV algorithms. - Analysing processed sensor data provided by Nissan. - Designing and carrying out controlled virtual reality studies of the targeted interaction scenarios, and analysing the collected data. - Applying, extending, and/or developing mathematical models of road user interactions to the targeted scenarios. - Investigating how to best integrate the mathematical models within the real-time perception and decision-making algorithms of an AV. These plans are flexible and will be agreed in collaboration between the PhD student, supervisors, and sponsor as the project unfolds.

For more information at to apply please refer to: 

https://phd.leeds.ac.uk/project/447-phd-studentship-sponsored-by-nissan-cognitive-models-of-human-traffic-interaction-for-real-time-behaviour-interpretation

Best,
Matteo