GUWAHATI: Researchers at the Indian Institute of Technology Guwahati have developed mathematical and machine learning models to analyse and predict traffic movement under heterogeneous Indian road conditions using drone-based data collected from a highway near Delhi. Led by Prof Akhilesh K. Maurya of the Department of Civil Engineering, the team tracked more than 36,000 vehicles and nearly nine million vehicle positions to study the relationship between traffic density and speed.
The study found that vehicle spacing changed with traffic conditions. Vehicles maintained larger gaps on relatively empty roads, moved closer together as traffic increased and again had larger gaps under heavily congested conditions. The researchers also observed that nearly 3,000 vehicles passed a point in each lane under full traffic flow, while density reached nearly 900 vehicles per kilometre across the road's entire width during traffic jams.
Using the dataset, the team tested six mathematical models and trained two machine learning models. Instead of predicting a single vehicle speed, the models estimated a range of possible speeds to account for the variability of traffic on Indian roads.
Prof Maurya said drone-based data collection enabled researchers to capture vehicle interactions and traffic patterns at high spatial and temporal resolution, helping address the limitations of conventional models developed for more orderly traffic conditions.
The researchers said the models could support traffic simulation, road and intersection design, congestion management, capacity assessment and intelligent transportation systems. The findings were published in the International Journal of Intelligent Transportation Systems Research in a paper co-authored by Prof Maurya and research scholars Shekhar Singh and Surya H. Ravikumar.
Also Read: Gauhati University signs MoUs to boost AI skills training and global study opportunities