PhD Position in Vehicle Sensor and Remote Sensing Analysis for Road Safety

This position focuses on leveraging vehicle sensors, remote sensing, and machine learning to support modern urban road safety analysis as part of a larger EU project. The project introduces a human-centred, evidence-based approach that integrates actual and perceived safety for pedestrians, cyclists, and micromobility users, moving beyond conventional crash-focused methods. It combines multi-source traffic, infrastructure, vehicle, and health data with immersive eXtended Reality (XR) experimentation and explainable Artificial Intelligence to analyse safety-critical situations. Large-scale pilots in four European cities validate methods in real traffic and support cross-city learning.

• Design and implement a software/tool architecture capable of processing multi-source vehicle sensor, camera, and remote sensing data streams.
• Train machine learning and computer vision models within the pipeline to automatically detect built-environment infrastructure characteristics and related factors affecting micromobility safety and comfort.
• Ensure the software accepts diverse inputs and generates structured, high-quality data outputs optimized for spatial mapping, risk diagnosis, and downstream predictive safety modelling.
• Validate and refine the pipeline using real-world data from the project’s pilot cities, working in close cooperation with international consortium partners.