Projects / AI Innovation Challenge

Rating Road Safety Through Machine Learning to Prevent Road Accidents

Project completed! Results attached!

Omdena Featured image

The International Road Assessment Programme (iRAP) is a registered charity established to help tackle the devastating social and economic cost of road crashes. The charity’s vision is for a world free of high-risk roads. In this project, more than 30 technology changemakers have built AI based solutions to increase road safety by mapping the crash risk on roads.

The Problem

Road crashes are the biggest killer of young people worldwide aged 5 to 29 years. The tragedy of death and injury impacts every community on earth. Based on current trends 500 million people will suffer life-changing brain, spinal, limb, and internal injuries between now and 2030. The economic cost will be more than US$25 trillion. Almost half of those killed will be vulnerable road users – motorcyclists, bicyclists, and pedestrians. Low-income and middle-income countries, where nine out of ten of the world’s road deaths occur, will suffer the most impact from this global road safety crisis.

iRAP is a registered charity established to help tackle the devastating social and economic cost of road crashes. The charity’s vision is for a world free of high-risk roads. iRAP already works with partners all around the world to apply the iRAP Global Standard in assessing the crash risk on a road and the associated Star Rating performance of road infrastructure.

Road Assessment Programme partnerships are now active in more than 100 countries worldwide. After assessing more than 1,000,000 km of roads, iRAP data helps make over US$75 billion of road investment safer. iRAP and partners are currently completing Star Rating assessments using specially commissioned video surveys or, where available and recent enough, Street View or similar public image data.

The Solutions: AI for Road Safety

Contributing to the solution in increasing road safety, more than 30 machine learning engineers, subject matter experts, and mentors collaborated as part of an Omdena challenge to work towards iRAP’s vision of “a world free of high-risk roads.”

The challenge revolves around the three main objectives listed below. These are under the Ai-RAP initiative that aims to accelerate road assessments with the help of big data and AI.

Three main objectives of the challenge:

  • Source geo-located crash data and produce iRAP Risk Maps of the historical crashes per kilometer
  • Source road attribute, traffic flow, and speed data and map the safety performance and Star Rating. This applies to more than 100 million km of road worldwide.
  • Produce repeatable road infrastructure key performance indicators that can form the basis of annual performance tracking

In this pipeline, we aim to contribute to the second objective by automatically sourcing the crucial component of vehicle count under the traffic or vehicle flow attribute using satellite imageries with the help of Artificial Intelligence.

The Data

iRAP partners worldwide have already collected more than 1 million km of road data over the last 20 years. A sample of that data has been presented in the charity’s Vaccines for Roads resource and Big Data tool (check this and this link).

The Impact

The United Nations Sustainable Development Goal is to halve road deaths and injuries by 2030. This includes targets to ensure all new roads are 3-star or better for all road users and that by 2030 more than 75% of travel is on 3-star or better roads.

iRAP estimates that achieving those targets will save 450,000 lives every year and prevent more than 100,000,000 deaths and injuries.  The burden of road trauma on individuals, families, communities, and health systems will be significantly reduced. This will result in more than $8 of savings from every $1 invested.

AI Road Safety

AI for Road Safety

Your benefits

Join a thriving AI community in 85 countries

Work with changemakers from around the world

Address a real-world problem with your skills

Build up your skill-set while setting the stage for a meaningful career


Good English

A good/very good grasp in computer science and/or mathematics

Student, (aspiring) data scientist, (senior) ML engineer, data engineer, or domain expert (no need for AI expertise)

Programming experience with C/C++, C#, Java, Python, Javascript or similar

Understanding of ML and Deep learning algorithms

This challenge has been hosted with our friends at

Application Form
media card
AI-Driven Temperature Analysis for Educational Environments in Tanzania
media card
CanopyWatch - Enhancing Deforestation Monitoring and Conservation in the Congo Basin using Machine Learning
media card
Streamline the Identification of Suitable Sites for Solar Panel Installations in UK

Become an Omdena Collaborator

media card
Visit the Omdena Collaborator Dashboard Learn More