PhD in Safety Gatekeeper for Runtime Safety Assurance of Autonomous Vehicles

Autonomous vehicles (AVs) are positioned to transform future transportation, yet their safe deployment remains a critical unsolved challenge. Despite intensive research since 2017, today’s AVs still rely on human supervision, largely due to the difficulty of validating the safety of systems built on complex AI components for perception, prediction, and control. These models struggle with diverse and previously unseen scenarios and provide limited guarantees under out-of-distribution inputs or uncertain training data. The rise of End-to-End learning and Vision-Language-Action further complicates assurance, as traditional interpretable module interfaces are replaced by latent representations that hinder modular testing. Addressing this gap requires new knowledge and methodologies that deliver quantitative, real-time safety guarantees and accountability for AI-driven decisions. By integrating safety gatekeepers that evaluate driving risk and intervening proactively, this research advances a timely and urgent frontier: safeguarded AI for autonomous driving. 

The successful candidate will be supervised by Dr. Pavlos Tafidis and Dr. Cheng Wang from both partner institutions resulting in a joint PhD degree from Heriot-Watt University and the University of Edinburgh. This allows gaining access to cutting-edge facilities and expertise in robotics, AI and autonomous systems in a collaborative research environment across University of Edinburgh and Heriot-Watt University. In addition, the candidate will work closely with industry partners who will provide datasets and a real autonomous driving platform. 

• Open to UK home status candidates only (EU applicants must have settled/pre-settled status or indefinite leave to remain and meet residency requirements.) a 2:1 undergraduate degree (or equivalent).

Candidate Profile: We seek a highly motivated individual with: 

• A strong background in intelligent transportation, smart mobility, robotics, machine learning or autonomous systems. 

• Excellent programming skills (Python, C++, or MATLAB). 

• Ability to work independently and collaboratively in a multidisciplinary team. How to Apply: Please send the following to Dr Pavlos Tafidis (pavlos.tafidis@ed.ac.uk) or Dr Cheng Wang (cheng.wang@hw.ac.uk) no later than 16th January 2026 (5 PM UK time): 

• CV 

• Cover letter outlining your motivation and relevant experience

the University’s English language requirements.

Applications are welcomed from self-funded students, or students who are applying for scholarships from the University of Edinburgh or elsewhere

Competition (EPSRC) funding may be available for an exceptional candidate. Link below for the further details.

Further information and other funding options.

Funding Details: 

• Duration: 3.5 years (42 months) 

• Start Date: Flexible, but no later than May 2027 

• Stipend: 10% above UKRI standard rate 

• Research & Training Grant: £3,500 total