Rouekela: A research team of National Institute of Technology – Rourkela (NIT-R) has secured a patent for an artificial intelligence (AI) model that optimises VANET (Vehicular Ad Hoc Network) system on how vehicles communicate with each other in future.
VANET is a self-organising wireless network, allowing nearby vehicles to share alerts, such as sudden braking or road obstacles, for better road safety, traffic management, and navigation. It provides real-time traffic updates, enhances emergency response, supports geolocation, and enables electronic payments.
Titled ‘Adaptive Contention Window Optimisation in VANETs using Multi-Agent Deep Reinforcement Learning for Enhanced Performance Model,’ the patent was secured by assistant professor Arun Kumar, Prof Bibhudatta Sahoo, and research graduate Lopamudra Hota from NIT-R’s computer science and engineering department.
Their AI model integrates multi-agent deep reinforcement learning with VANETs and addresses any glitch in an overloaded system, leading to delays when multiple vehicles simultaneously send out messages, by giving priority to messages that are important and time-sensitive.
The developed model ensures that even in busy conditions, the right message reaches at the right time, which can support safer mobility.
Prof Sahoo emphasized that the patent advances vehicle-to-vehicle interaction by mitigating network congestion, laying the foundation for safer and more efficient traffic systems in India.
Prof Arun noted that India reported 4.80 lakh road accidents and 1.72 lakh deaths in 2023, many preventable with advanced technology. “Our model supports safer roads and smarter cities, aligning with Innovate in India and Make in India, paving the way for autonomous vehicles,” he added.