Verifiable, Safe and Interpretable Multimodal Large Language Model Control System Design for Vehicle Environment Understanding

This PhD project investigates how multimodal large language models can be used to design an AI-guided predictive control framework for vehicle applications where multiple control systems must work together safely and effectively. The research centres on the Run Dry Traction System (RDTS), an active safety system that removes water from in front of each tyre and supports torque vectoring to improve safety and performance on wet road surfaces. The project will focus on: designing an AI and model predictive control framework for vehicle systems managing interaction between RDTS and existing vehicle safety control systems using multimodal sensor fusion to detect water level and road conditions refining MLLM-based control and activation strategies improving cornering, traction, stability, and agility in automated vehicles The first two years will be based at Coventry University in the UK and focus on vehicle dynamics modelling and control interaction. The final two years will be based at GITAM in India and focus on MLLM agents and sensor fusion for RDTS activation.

Minimum 2:1 first degree in a relevant discipline
Minimum 60% in the project element or equivalent
Minimum 60% overall module average
Ability to complete the PhD within 3.5 years
IELTS 7.0 overall with minimum 6.5 in each component

For more information, contact:

csx259@coventry.ac.uk

Applications require:

full supporting documentation
covering letter
2000-word supporting statement explaining how your expertise and interests match the project