Predictive Path Analyser For Addressing Road Safety, Traffic Congestion

Predictive Path Analyser For Addressing Road Safety, Traffic Congestion

Students' Corner October 2019 Predictive Path Analyser Addressing Road Safety
{data-social}

Team PathPredictor's solution determines road traffic density using a radar system and map data from the internet, and identifies the priority vehicles present on the road

Road safety and traffic congestion are huge concerns in India. And to address these aspects, Team PathPredictor, led by Animesh Srivastava from the National Institute of Technology, Hamirpur, Himachal Pradesh, worked on developing a Predictive Path Analyser – an enrouting & accident prevention self-learning system.

Sample this: over 150,000 people are killed in India each year in road accidents that translate into 400 deaths daily; one person dies every four minutes. Bear in mind the fact that these are just official numbers, and that unaccounted deaths could be much higher. Along with road-accident-related deaths, traffic jams on roads is another big issue. In fact, traffic congestion, especially in metropolitan cities serves as a prominent impediment to the operation of priority vehicles such as ambulance, fire truck, police vehicles, etc.

Team PathPredictor won the most popular award at the 2018 KPIT Sparkle Contest

PROBLEM STATEMENT

There is a strong need for an intelligent system to ameliorate the on-road traffic jam scenario with an aim to provide data about the most advantageous paths available to a priority vehicle, and thus help realise its services in an efficient manner. Predictive Path Analyser helps in extracting and analysing traffic density statistics and ensure it is available for priority vehicles and end-users, which will help choose the right route, in terms of safety and time.

In developing the Predictive Path Analyser, the team bore in mind the ever-increasing number of road accidents and delays incurred for priority vehicles. The focus of the Predictive Path Analyser is to tackle the time factor in saving a patient’s life and reduce the fire death rate. Team PathPredictor worked on developing a software for vehicle users to facilitate easy interpretation of road traffic data, aided through visualisation of the same. The software presents analysis data for road conditions along with the re-routing data so that vehicle users can safely reach their destination, clearing road traffic and providing way for priority vehicles.

A huge number of accidents are also reported from hilly terrains, particularly in low-light conditions due to blind turns and bad road conditions. The team claims that Predictive Path Analyser can help mitigate such accidents. This solution is an important step towards providing an optimal route for less hindrance-based movement.

Predictive Path Analyser helps in extracting and analysing traffic density statistics and ensure it is available for priority vehicles and end-users, which will help choose the right route, in terms of safety and time

SOLUTION

The proposed solution determines road traffic density using a radar system and map data from the internet, and identifies the priority vehicles present on the road. The team claims to have developed an efficient path determination algorithm, leveraging machine learning, and thus providing alternate routes to road users. Further, the Predictive Path Analyser also monitors crevices and bumps using accelerometer and gyroscope sensors that help prevent road accidents as well as reduce delays in reaching destinations. The data collected from the roads is used to monitor road conditions and help alert the user about the road conditions.

Additionally, Team PathPredictor has developed an algorithm that focusses on providing information about the best as well as the worst route for an emergency or priority vehicle. The application developed using the algorithm also focusses on creating mass awareness about road safety. This Predictive Path Analyser application can be developed for iOS as well as Android-based devices for easy access and use, Animesh said.

CONCLUSION

The potential customers, who require better road and traffic statistics, only have to possess a smartphone with internet connection – a fairly low implementation cost. The cost of development is low, and depends on testing and simulation of the application in real life. The application can be further integrated with more features like air pollution monitoring, smog particle count and real-time SOS help.

At the 2018 KPIT Sparkle Contest, Team PathPredictor won the most popular award and earned high scores from the jury for its Predictive Path Analyser. The Animesh Srivastava-led team was selected based on public voting from across the country and won a cash prize of ` 1 lakh. Animesh’s team comprised of B Jatin Rao and Aditya Arora, and they received good support from fellow students at NIT, Hamirpur.

TEXT: Suhrid Barua