Verifiable, safe and interpretable Multimodal Large Language Model control system design to accommodate interaction between multiple control systems exploiting vehicle environment understanding
Eligibility: UK/International (including EU) graduates with the required entry requirements
Duration: Full-Time – between three and three and a half years fixed term
Application deadline: 1st May
Interview date: Will be confirmed to shortlisted candidates
Start date: September 2026
For further details contact: Associate Professor Olivier Haas
Introduction
The increased task complexity associated with vehicle automation requires managing the increased number of interactions between multiple sensors and control systems that exploit their information. Each control system will have its own objectives and constraints. It is necessary to ensure that they all work cooperatively to manage potential conflicts.
Multimodal Large Language Models (MLLMs) have proven useful in automotive applications to support environmental detection and interpretation (e.g., rain and floods). MLLMs can also support the design and tuning of control systems and manage the complexity of finding the optimal trade-off among multiple objectives and constraints.
Run Dry Traction System (RDTS) is a new active safety system that removes water from the front of each tyre and can supplement the torque-vectoring capabilities intrinsic to new electric and automated vehicles. It is a novel and disruptive technology with the potential to significantly improve safety when vehicles operate on wet road surfaces. In addition, it can facilitate torque vectoring and improve vehicle performance and safety.
Project details
This project will exploit MLLM capabilities to design a new Artificial Intelligence (AI) guided predictive control framework for vehicle applications.
The framework will be applied to the Run Dry Traction System (RDTS) which will need to interact with other active safety control systems on the vehicle.
In the UK, the research will focus on the engineering problem associated with deploying the RDTS system to control the activation of the Run Dry Traction System (RDTS) in cooperation with current dynamic stability control systems.
In India, the research will focus on the multi-sensor fusion to detect water level and road condition warrinting the RDTS activation and refine the MLLM based framework.
Background Information
Vehicle dynamic control systems operate the stability, traction, brake and torque distribution systems. These controllers exploit information from the vehicle and its environment, combined by sensor fusion algorithms and contribute to high level controllers automating the vehicle operation. The increasingly complex environment combined with the development of new features and systems is further increasing the overall control system and associated actuator architecture complexity.
Large Language Model (LLM) agents are currently being developed to tune control loops, resulting in faster development cycles, reducing the need for experts, cross-domain awareness and the ability to produce documentation and high-level reasoning whilst delivering low-level implementation. Multimodal LLM (MLLM) can further understand multiple data types, which is essential in designing hierarchical control systems that need to manage the trade-off between multiple objectives and constraints.
This PhD project aims to exploit MLLM capabilities to design a generalisable framework combining Artificial Intelligence (AI) and model predictive control with application to the activation and control of the Run Dry Traction System (RDTS), which can remove water in front of each tyre.
In the first two years at Coventry University, UK, the research will focus on vehicle dynamic modelling and benchmarking of dynamic stability control systems. It will investigate the interactions between the RDTS controller and existing dynamic control safety systems. Engineering knowledge will be applied to begin the framework for designing an overall controller to improve the vehicle's cornering, traction, stability, and agility.
Over the last two years, at GITAM, Visakhapatnam Campus, India, the work will focus on the design of MLLM agents and sensor fusion to ensure that RDTS is activated only when required.
Dr Olivier Haas – Coventry University https://pureportal.coventry.ac.uk/en/persons/o-c-l-haas/ is an Associate Professor and the Intelligent Mobility and Control team lead. He is one of the most successful directors of studies in the University with 30+ PhD completions. His research interests include Model Predictive Control, AI, and Machine Vision, with applications to automotive systems.
Dr Qian Lu is an Assistant Professor in Connected and Autonomous Vehicles, within the Intelligent Mobility and Control team. Her research interests include generative AI and control with application to advanced driver assistance systems.
Professor Mike Blundell is well known for his work in vehicle dynamics and the use of multi-body systems (MBS) software. As the RDTS inventor with experience developing and testing an RDTS prototype at HORIBA MIRA, he will provide expert knowledge on RDTS and vehicle dynamics.
Professor Bharani Chandra is Director of the GITAM School of Core Engineering and co-founder at CtoSKYAI Pvt Ltd. His areas of research include control systems design and state estimation with applications including autonomous vehicle systems.
Funding
Tuition fees and bursary
Studentships include one return economy airfare to GITAM/Coventry University, visa and overseas healthcare to cover the mobility period.
Benefits
The successful candidate will receive comprehensive research training including technical, personal and professional skills. All researchers at Coventry University (from PhD to Professor) are part of the Doctoral and Researcher College, which provides support with high-quality training and career development activities.
Entry requirements
- A minimum of a 2:1 first degree in a relevant discipline/subject area with a minimum 60% mark in the project element or equivalent with a minimum 60% overall module average.
PLUS
- The potential to engage in innovative research and to complete the PhD within 3.5 years.
- A minimum of English language proficiency (IELTS academic overall minimum score of 7.0 with a minimum of 6.5 in each component).
How to apply
To find out more about the project, please contact:
Associate Professor Olivier Haas
All applications require full supporting documentation, a covering letter, plus a 2000-word supporting statement showing how the applicant’s expertise and interests are relevant to the project.
Apply to Coventry University
