Argus News - NIT Rourkela Patents AI Model to Boost Road Safety

Rourkela, Sep 29: In a significant stride toward smarter mobility, researchers at the National Institute of Technology (NIT) Rourkela have patented an advanced artificial intelligence model designed to enhance vehicle-to-vehicle communication and road safety. The innovation targets a key challenge in Vehicular Ad-Hoc Networks (VANETs), where message congestion can lead to delays and compromised safety.

Smarter Communication for Safer Roads
The patented model, titled “Adaptive Contention Window Optimisation in VANETs using Multi-Agent Deep Reinforcement Learning for Enhanced Performance Model”, was developed by Dr. Arun Kumar, Prof. Bibhudatta Sahoo, and Dr. Lopamudra Hota from the Department of Computer Science & Engineering.

VANETs allow vehicles in close proximity to exchange real-time data, such as alerts about sudden braking or road hazards. However, simultaneous messaging from multiple vehicles often causes network congestion, leading to delays or lost signals.

AI-Powered Coordination
To address this, the NIT team introduced a multi-agent deep reinforcement learning system. This AI-driven approach enables vehicles to adaptively schedule their messages based on surrounding traffic behavior, ensuring that time-sensitive alerts are prioritized and transmitted without delay.

Dr. Arun Kumar emphasized the urgency of such innovations, citing India’s 2023 road accident statistics: “Around 480,000 accidents and 172,000 deaths were reported. Many could be prevented using modern technologies.”

Applications and Future Vision
The model supports a wide range of applications including:

-Electronic brake lights

-Platooning for coordinated driving

-Real-time traffic updates

-Emergency alerts

-On-the-go services and smart city integration

Prof. Bibhudatta Sahoo added, “This patent lays the groundwork for efficient traffic management and future autonomous vehicles. We welcome collaboration from institutions nationwide.”