Potential improvements in rail safety will be investigated as part of a research study pitting an artificial intelligence tool against the human brain’s ability to track and process hazards.
The Optimus prototype — installed at a Network Rail-operated level crossing near Cheltenham in England earlier this year — uses machine learning and an AI-based object detection system to identify and quantify different types of traffic.
It will allow researchers from the University of Leicester to compare both the accuracy and speed of its detection capability to a human completing the same task.
It is believed that the study is the first in comparing humans and AI for the task of “visual census” in this way.
Researchers in the School of Computing and Mathematical Sciences have combined efforts with visual perception experts from Leicester’s Department of Neuroscience, Psychology and Behaviour to shape the study, which will track participants’ eye movements when faced with several fast-moving hazards simultaneously.
The research will be used to inform the next generation of a rail safety devices.
“The question was asked on the project ‘How good is good?’, in relation to the performance of the Optimus system at the crossing,” said George Leete, KTP research associate within the AI, Data Analytics and Modelling Centre at the University of Leicester.
“I am confident that the results of this study will lend us some valuable insight into this question, firstly how good humans are at the current task and, secondly whether our current system holds up to human standards,” said Mr Leete, who leads development of the machine-learning aspect of the project.
“We believe this is the first time an AI-based system will be validated against a standard in this way, opening the door for other systems to be validated via a similar method.”
There are about 6,000 level crossings in the UK, said the Network Rail, which is responsible for the country’s rail infrastructure. Figures for 2019 and 2020 show that there were 316 near misses with pedestrians on UK level crossings, and two pedestrian fatalities.
The Optimus prototype was installed at a site on the Cross Country Route in January 2022 and has already identified hundreds of thousands of movements on the crossing, including movements by pedestrians, cyclists and other road traffic.