Academics at the Heriot-Watt University School of Engineering and Physical Sciences have begun using advanced machine learning technology to improve the performance of autonomous vehicle testing simulators.
Technology has been applied to existing human driving behavioural models in order to eliminate existing system limitations through the use of additional data to improve both the performance and design of next generation simulators.

Research thus far aims to focus on interactions and human-like decisions in order to provide prospective developers with an artificial intelligence (AI) tool capable of simulating a larger range of realistic driving situations and scenarios.
With aims to develop high-fidelity, realistic simulators, the team hope to accelerate the overall development and iteration of autonomous vehicle technology, as well as aid in the definition of testing frameworks for use by policymakers.
The programme has been funded with a grant of just under 200,000 GBP from the European Commission’s Horizon Europe Framework Programme, with initial findings expected to be published before 2026.
Dr Cheng Wang, Assistant Professor in Robotics, said:Autonomous vehicles (AVs) have the potential to significantly enhance the safety and sustainability of transportation in the future.
While substantial progress has already been made, the safety assessment of AVs remains a serious challenge, delaying their widespread adoption.
Typically, the safety of AVs is measured by comparing them to the performance of human drivers, however this requires hundreds of millions of miles of real-world testing, a process inherently fraught with unpredictable risks.
We’ll be delving into advanced machine learning, using data to more accurately and realistically model human driving behaviour. We hope our work will ultimately help speed up the development and safety of autonomous vehicles by improving the simulation systems.