Tags:Science & Tech
For Prelims: Machine Learning (ML), Drone (Unmanned Aerial Vehicle - UAV), Traffic Congestion, Total Models Developed, Vehicles Studied, Vehicle Movements Analysed, Data Collection Method, Nodal Body
For Mains: Drone-based traffic monitoring, AI for urban governance, Mixed traffic conditions in India
Source:
Hindustan
5 facts for one-day exams
SSC, railway, police and teaching exams ask direct questions like these.
- IIT Guwahati developed 8 traffic prediction models using drones and AI.
- The models include 6 mathematical and 2 machine learning models.
- Data was collected from over 36,000 vehicles on a Delhi highway.
- The study analysed 90 lakh (9 million) vehicle movements.
- The models are designed for India's mixed traffic conditions.
Why in the news?
IIT Guwahati researchers have developed eight models (six mathematical and two machine learning) using drone footage and AI to predict traffic congestion on Indian roads. The models, trained on over 36,000 vehicles and 90 lakh movement data points from a Delhi highway, are designed for India's mixed traffic conditions and can forecast jams, assist traffic police, and improve road design.
Summary
- IIT Guwahati researchers developed eight traffic prediction models (six mathematical, two machine learning/AI) using drone video data from a Delhi highway.
- The study analysed over 36,000 vehicles and 90 lakh (9 million) vehicle movements to understand Indian mixed-traffic behaviour.
- Models can predict where and when traffic jams may form, helping traffic police, road design, and emergency vehicle routing.
Highlights
Drone Data Collection
Researchers used drones to record vehicle movements on a Delhi highway, capturing data from over 36,000 vehicles and 9 million movement points.
AI and Mathematical Models
Six mathematical and two machine learning models were developed to predict traffic speed and congestion patterns on Indian roads.
India-Specific Traffic Models
Unlike models from other countries, these are designed for India's mixed traffic conditions with diverse vehicles like bikes, scooters, auto-rickshaws, cars, and buses on the same road.
Background
Traffic congestion is a major urban challenge in India, leading to economic losses, fuel wastage, and increased pollution. Existing traffic models are often based on homogeneous traffic conditions common in developed countries, where vehicles move in a more uniform manner. India's roads have mixed traffic with diverse vehicle types, speeds, and driving behaviours, making prediction difficult.
Key Analysis and Impact
The IIT Guwahati models use real-world Indian traffic data captured via drones, making them more accurate for local conditions. The AI models can learn from historical data to forecast congestion, enabling proactive traffic management. This can help traffic police deploy resources, optimise signal timings, and improve emergency vehicle access, potentially reducing travel time and fuel consumption.
Way Forward
These models need to be tested on more roads across different Indian cities and integrated with existing traffic management systems. Scaling up with real-time data feeds from traffic cameras and sensors could create a national traffic prediction system. Further research could refine the models for specific road types, weather conditions, and festival seasons.
Background GK
Textbook concepts you need to understand this news
- Machine Learning (ML)
- A subset of artificial intelligence (AI) where algorithms learn patterns from data without being explicitly programmed, enabling predictions or decisions.
- Drone (Unmanned Aerial Vehicle - UAV)
- An aircraft without a human pilot onboard, controlled remotely or autonomously, used for surveillance, data collection, and delivery.
- Traffic Congestion
- A condition on road networks where vehicle density exceeds capacity, leading to slower speeds, longer trip times, and queuing.
Facts to remember
Tap a card to see the answer. Try to recall it first.
How exams will ask this
- Data: IIT Guwahati developed 8 models (6 mathematical + 2 ML); studied 36,000+ vehicles and 90 lakh movements.
Don't make this mistake!
Examiners may ask for the total number of models (8) and confuse the split (6+2) or swap the number of vehicles (36,000) with the number of movements (90 lakh). They may also state the models were developed by IIT Delhi or IIT Kanpur instead of IIT Guwahati.
Write these in mains answers
- Drone-based traffic monitoring, AI for urban governance, Mixed traffic conditions in India
Stuck? Get it in one tap
Scientists at IIT Guwahati used drones to film traffic on a highway near Delhi. They studied how different vehicles like bikes, cars, and buses move together. Using this data, they created eight computer models that can predict where and when traffic jams might happen. These models are special because they work for Indian roads where many types of vehicles travel together.
- IIT Guwahati developed 8 traffic prediction models using drones and AI.
- The models include 6 mathematical and 2 machine learning models.
- Data was collected from over 36,000 vehicles on a Delhi highway.
- The study analysed 90 lakh (9 million) vehicle movements.
- The models are designed for India's mixed traffic conditions.
- Science and Technology: Artificial Intelligence and Machine Learning
- Governance: Urban Infrastructure and Smart Cities
Related concepts: Machine Learning (ML), Drone (Unmanned Aerial Vehicle - UAV), Traffic Congestion
Practice questions
Pick an answer, then read the explanation.
- Q1Easy
Which Indian institute developed the new traffic prediction models using drones and AI?
Explanation: The news specifically states that IIT Guwahati's scientists developed the eight models. IIT Delhi, IIT Bombay, and IIT Kanpur are other premier IITs but were not involved in this research.
- Q2Hard
Consider the following statements regarding the traffic prediction models developed by IIT Guwahati:
- 1.The models are based on data collected from drones flying over a highway near Delhi.
- 2.A total of eight models were developed, including six machine learning models.
- 3.The models are designed specifically for India's mixed traffic conditions.
Which of the statements given above is/are correct?
Explanation: Statement 1 is correct: data was collected via drones on a Delhi highway. Statement 2 is incorrect: the eight models include six mathematical and two machine learning models, not six machine learning. Statement 3 is correct: the models are tailored for India's mixed traffic. Hence option C is correct.
- Q3Medium
How many vehicle movements were analysed by the IIT Guwahati researchers for developing the traffic prediction models?
Explanation: Researchers studied the movement of over 36,000 vehicles and their approximately 90 lakh (9 million) activities. 36,000 is the number of vehicles, not movements. 9 lakh is too low, and 90,000 is also incorrect.
- Q4Medium
What is the primary advantage of the traffic prediction models developed by IIT Guwahati over existing models?
Explanation: Existing models are for countries with homogeneous traffic, while India has mixed traffic with bikes, cars, buses, and pedestrians on the same road. These new models account for that.
Frequently asked questions
1. Who developed the new traffic prediction models using drones and AI?
Researchers at the Indian Institute of Technology Guwahati (IIT Guwahati) developed the models.
2. How many models were developed and what are they based on?
Eight models were developed: six mathematical models and two machine learning (AI) models, based on drone video footage of a highway near Delhi.
3. How much data was used to train these models?
The study analysed over 36,000 vehicles and approximately 90 lakh (9 million) vehicle movements.
All sources
- Original5 November 2026 · Read story
