NIT Rourkela develops AI model to improve road safety

NEW DELHI: National Institute of Technology (NIT) researchers have developed an AI model that aims to improve how vehicles communicate with each other in vehicular ad-hoc networks (VANETs), according to officials. The patented model, called 'Adaptive Contention Window Optimisation in VANETs using Multi-Agent Deep Reinforcement Learning for Enhanced Performance Model', focuses on addressing a key challenge in vehicular communication systems.

According to officials, the concept behind VANETs is that vehicles which are in proximity to each other will in the future be able to communicate directly with each other. Consider a car warning other vehicles about precipitous braking or a sudden obstacle on the road. Such communication aids driving, automated traffic systems and even provides assistance to emergency services.

However, when multiple vehicles are contemporaneously sending out messages, the system is bound to face vehicle overcrowding. This congestion leads to delays or lost messages, which directly compromise the viable functionality of such systems.

Also read NIT Rourkela Placements 2025: 1,274 job offers, highest salary Rs 62.44 LPA despite global recession

Use of artificial intelligence

"We have suggested a solution to this issue involving the use of artificial intelligence. Their model utilises a system known as multi-agent deep reinforcement learning. In layman's terms, it enables each vehicle to stagger the time of its messages depending on the actions of other vehicles.

Instead of communications competing, the vehicle's system learns to sequence and give priority to messages that are time sensitive," said Arun Kumar, Assistant Professor, NIT Rourkela . This adaptive adjustment reduces the chances of delay and helps ensure that important alerts are transmitted reliably.

"In 2023, India reported around 4,80,000 road accidents and around 1,72,000 deaths, many of which could be prevented using modern technologies. Our work is a step towards building safer roads and smarter cities. "We envision a near future where autonomous vehicles become a reality in India, and this patent is a small step in that direction, driving the spirit of Innovate in India and Make in India," he added.

Also read NIT Rourkela Placements 2025: 1,274 job offers, highest salary Rs 62.44 LPA despite global recession

AI model for busy situations

The developed model ensures that even in busy conditions, the right message reaches at the right time, which can support safer mobility. This approach represents a step toward future transportation systems where vehicles can coordinate in real time. VANETs use case scenario including electronic brake lights that notify drivers of braking scenarios that are not in their line of sight and platooning that enables cars to closely tail a lead car through distributed acceleration and steering control data.

Kumar explained that enhancing navigation systems, VANETs can obtain, process and provide instantaneous information to users on current traffic conditions of roads, improve emergency response systems by rapid broadcast of crucial information, support geolocation systems pedalling electronic payments on-site for nearby retail stores and on the move facilities, such as restaurants, and provide remote access to real-time data for electronic toll collection systems.

"The patent represents a practical step towards preparing India's road system for vehicle-to-vehicle communication. By addressing potential congestion in VANETs and providing a model for adaptive, coordinated communication, the findings lay the groundwork for safer and more efficient traffic management.

"We invite researchers across institutions to join our efforts in making future autonomous vehicles and collaborate with our research lab at NIT Rourkela," said Bibhudatta Sahoo, Professor, NIT Rourkela.