Accelerated and intelligent RAP data collection
Improving road safety performance is a priority for all levels of government in Australia. The Australian Road Assessment Program’s (AusRAP) data is currently used to inform policy, investment and design decisions across the country.
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 AusRAP information.
The AiRAP (accelerated and intelligent RAP data collection) project will deliver the digital data specifications, compliance, and quality assurance processes to enable a competitive data marketplace to be developed.
Pilot AiRAP assessments will be conducted on a sample of attributes across local, state and national highway roads to prove the methodologies and provide the framework and confidence levels for the digital data generated.
Australian transport ministers have set star rating policy targets and risk assessment requirements for all states and territories as part of the National Road Safety Action Plan 2018-2020. More than 100,000 kilometres of AusRAP Star Rating assessments have already been undertaken across the country that includes over 50 attributes coded to a global standard every 100 metres.
These assessments have traditionally been collected using video data and manual human coding of attributes. The accelerated and intelligent collection of AiRAP provides a unique and significant opportunity for the new Office of Road Safety and national, state, and local governments to more cost effectively manage their safety performance and support achievement of the National Action Plan Targets.
The ability to performance track KPIs over time (and retrospectively) in a repeatable and standardised manner to a global standard will be highly valuable to all authorities.
The project will deliver 20,000 kilometres of road attribute data for the state road network in New South Wales using TomTom’s MN-R 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. The initiative will ultimately open up existing and emerging data sources for network-level road safety assessments throughout Australia and worldwide.
As part of the Australian partnership with Austroads, AAA, ARRB and iRAP, the project supports the AusRAP outcomes as well as local derivatives. The accelerated and intelligent collection and coding of road attribute data to a common global standard has the potential to:
- reduce the time and effort required to undertake road safety assessments
- reduce the costs and improve the accuracy and frequency of collection for performance tracking and KPI monitoring purposes
- contribute to improvements in road safety and saving lives.
The project’s objectives are to:
- Develop AiRAP Framework
- Prove AiRAP Framework and scalability
- Prove independent feature extraction
- Create an AiRAP supplier map and estimate efficiency savings
- Recommend Phase 2 priorities
UPDATE: May 2023 – Final report available
This project is now completed, and a wrap-up article has been published, which states the case for the technology and its application to road safety, main project outcomes, and details of a proposed phase 2 of the project. It also includes a downloadable copy of the final report. Find the article, and the report at: AiRAP automation for Australian road safety