Transport for NSW launch AI data mining project to rate roads for safety & lower death toll

301
Media Release

In a world where people  are embracing domestic road trips due to the impact of the Covid-19 pandemic, a recently launched initiative in NSW aims to use artifical intelligence (AI) to make our road travel safer.

Transport for NSW (TfNSW) has established a timely project using cutting-edge AI to convert raw data from our roads into an international standard five-star ratings systems, as well as to potentially develop predictive algorithms for fatality and injury outcomes.

The project is supported by the iMOVE Cooperative Research Centre and is being led by (TfNSW) and partners, including the University of Technology Sydney (UTS), the International Road Assessment Programme (iRAP) and geospatial data experts Anditi.

The project will deliver up usable data for 20,000km of NSW roads to the state government using TomTom’s MN-R next generation map data, as well as prove feature extraction techniques and machine learning for LiDAR data.

The project aims to prove rapid, scalable and repeatable methods for road data extraction as part of iRAP’s global ‘AiRAP‘ initiative (accelerated and intelligent RAP data collection). The initiative will ultimately open up existing and emerging data sources for network-level road safety assessments throughout Australia and the world.

iRAP is a registered charity dedicated to saving lives by eliminating high risk roads throughout the world. It assesses roads all over the world and aims to significantly reduce road casualties by improving the safety of road infrastructure.

The Australian Government is linking infrastructure funding to “measurable improvements in safety” and state agencies are setting network and project-level star ratings before greenlighting public spending.

The 2018-2020 National Road Safety Action Plan set targets for 90% of travel on national highways and 80% of travel on state highways to meet a 3-star or better safety standard.

To date, more than 280,000km of Australian roads have been star rated.

The assessments involved identifying and recording more than 50 road attributes to the iRAP global standard every 100 metres. However, these assessments have been painstakingly done using video survey footage and manual recording methods.

IRAP’s Global Innovation Manager & Cities Specialist Monica Olyslagers is the project manager.

She said the use of AI and machine-learning would significantly speed up the task of star-rating roads to enhance public safety.

“Raising the standard of the world’s roads to a 3-star or better standard for all road users will help to focus policy and investment. With crash costs typically halving with each incremental improvement in star rating the potential for 3-star or better roads to save lives is significant,” Olyslagers said.

In recent years, between 1,100 and 1,200 people lose their lives to road crashes on Australian roads each year, and more than 40,000 suffer lifelong, debilitating injuries.

“The use of artificial intelligence and machine learning techniques to collect the data has potential to reduce costs and increase the frequency and accuracy of data,” Olyslagers said.

“Making faster and more affordable data collection possible means that safety assessments can be done on an annual basis across the whole road network.”

IMOVE CRC managing director Ian Christensen said improving road safety performance is a priority for all levels of government in Australia.

“Using technologies such as AI to enhance in our suite of safety policy tools is a great step forward. These powerful and insightful tools can inform sound investment by government that saves lives and unlock significant benefits to families, communities, business and health systems through reduced road trauma,” he said.