For outstanding, high-quality research and future contribution to our profession.
Each year, the SIDRA SOLUTIONS Postgraduate Award is given to the most outstanding postgraduate student from a tertiary institution in Australia or New Zealand, based on the quality of his or her research and potential to make a significant contribution to the traffic and transport engineering profession.
ITE-ANZ and SIDRA SOLUTIONS are proud partners in this award.
Winner 2025: ZHENG XU
Congratulations to Zheng Xu of the Monash Institute of Transport Studies, Monash University for winning the 2025 SIDRA SOLUTIONS Postgraduate Award for his thesis titled 'Human-Centric Analysis of Critical Scenarios and Operational Efficiency during High-level Autonomous Driving'.
Summary of Research
Zheng's PhD research addresses critical safety challenges in autonomous driving systems (ADS) by focusing on human-ADS interaction. Using immersive VR-enabled simulation and human-in-the-loop experiments, his research demonstrates that human intervention behaviours during autonomous driving can reveal safety gaps in ADS, and that generating intervention-involved scenarios is essential for effective ADS training and efficient operation of high-level autonomous vehicles.
The research makes three key contributions: a novel methodology for generating safety-critical scenarios that are rare in naturalistic driving; a framework for analysing the complexity of human riders' interventions; and approaches for enhancing ADS automation and resilience from a human-centric perspective. Through a scene-scenario-situation structure integrating CARLA and VISSIM co-simulation, the research measures scenario criticality and quantifies the discrepancy between ADS and human riders, characterises and predicts intervention behaviour, and develops a calibrated platform to generate critical data for ADS training.
The findings have broad transferability, informing automotive manufacturers' testing processes, human-AV interface design, regulatory frameworks, and the development of more context-aware autonomous vehicles.
Here are two papers by Zheng: