Back to Projects

Team Name:

Sirius


Team Members:


Evidence of Work

Roadcrash Guard

Project Info

Team Name


Sirius


Team Members


Vinod , Uma

Project Description


RoadCrash Guard is a responsive AI based solution designed to predict high-risk crash hotspot locations using Victoria Crash data. The solution will predominantly use three channels.
• A website that can provide a summary of road crashes on the Victoria map, with filters for various parameters involved in the crash.
• An end user mobile application, that can notify and warn users of a road crash risk based on their real time location.

• An agent that generates comprehensive and concise executive briefs that summarize key findings, insights, and recommendations.
Key Features:
• Crash Hotspot Prediction: Identifies high-risk areas for crashes.
• Mobile App: To enhance Road Safety for end users.
• Executive Briefs: Provides summaries of key findings, insights, and actionable recommendations.
Recommendation Analysis Parameters:
• Latitude and Longitude of the accident spot
• Road/Street name
• Road condition
• Intersection type
• Atmospheric condition
• Vehicle type and age
Our Goals for RoadCrash Guard:
• Develop a web page to pinpoint all crash hotspots in Victoria with filters to indicate the vehicle type, road conditions and atmospheric conditions.
• Generate comprehensive and concise executive briefs summarizing our findings, insights, and recommendations for government officials.
• Create a mobile application to enhance road safety for users such as pedestrians, bicyclists, motorcyclists, drivers, and passengers, based on their current location.
• The key differentiator is the feeding in of recent data on a weekly basis, to enhance the prediction capability of the AI system.


Data Story


https://discover.data.vic.gov.au/dataset/victoria-road-crash-data

Road crash data reflects the economic, personal and social wellbeing of the society. We leveraged the Victorian Road Crash data geojson to obtain the coordinates of the road crash spots. This data is leveraged to pinpoint high risk areas. Using this data along with the Road condition, Vehicle Age and Atmospheric conditions allowed us to understand what triggers accidents and the nature of fatality.

There are certain intersection points which seems to have high road crashes irrespective of road or atmospheric conditions. These are due to narrow roads or unexpected objects blocking roads.
Yet another data leveraged(but not used due to lack of time) was the nature of accident itself. Head on car collisions are mainly caused by narrow road allowing traffic in both directions or blind spots. These will require decisions to allow these roads to support uni directional traffic. Yet another are crashes with stationary objects which would need the council to ensure roads are cleared of any possible objections to traffic.
Overall, the data gave us good insight into probable infrastructure improvements to help reduce accidents.


Evidence of Work

Video

Homepage

Project Image

Team DataSets

This team does not currently have any datasets.

Challenge Entries

AI applications using Open Road Crash data

How might we leverage road crash statistics and multi-agent AI-based web applications to enhance road safety and inform policy making?

Go to Challenge | 13 teams have entered this challenge.