AI DRIVEN TRAFFIC LIGHT MANAGEMENT SYSTEM
- 1. SREE BUDDHA COLLEGE OF ENGINEERING(AUTONOMOUS),PATTOOR DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING (AI&ML) 2024-2025 AI-DRIVEN TRAFFIC LIGHT MANAGEMENT SYSTEM GUIDED BY LEKSHMI V. S. ASST.PROFESSOR CSE(AI&ML) PRESENTED BY NANDA B. R. (SBC22AM038) NANDANA S. NAIR (SBC22AM039) PARVATHY S. KAMAL (SBC22AM045) THEJAS S. (SBC22AM055)
- 2. CONTENTS • INTRODUCTION • LITERATURE REVIEW • PROBLEM DESCRIPTION • OBJECTIVE • PROPOSED METHODOLOGY • WORKFLOW DIAGRAM • RESULT • FUTURE SCOPE • CONCLUSION
- 3. INTRODUCTION Traditional traffic signals follow fixed timings , causing congestion and delays. Manual intervention is inefficient and labour intensive. AI-driven traffic lights detect real-time traffic density and adjust signals dynamically This improves traffic flow ,reduces congestion , and enhances road safety AI-DRIVEN TRAFFIC LIGHT MANAGEMENT SYSTEM
- 5. SL. no TITLE AUTHOR(S)& YEAR SUMMARY LIMITATION 1. AI-Based Adaptive Traffic Control for Congestion Mitigation. Dr G.Vishnu Priya , Shyamala Gouri (2024) Uses real-time data to cut congestion and prioritize emergency vehicles. High cost and data needs limit use in smaller cities. 2. AI Adaptive Traffic Signal Control for emission reduction. C . Ashok kumar , N. Anuradha (2024) AI traffic control using Deep Q- Learning to reduce emissions. Data-dependent , and hard to scale in real world. 3. YOLO-based Traffic Signal Optimization for Smart Flow Control Joesam Dinesh C , Dr . Sibi Amaran(2024) YOLO-based system adjusts signals via live feeds to cut traffic and pollution. Lacks focus on scaling, occlusion, and emergencies. 4. AI-Driven Traffic Lights Denslin Nunes, Meet Satra(2023) AI traffic lights use real-time image data Privacy , setup and maintenance make
- 6. SL. no TITLE AUTHOR(S)& YEAR SUMMARY LIMITATION 5. Smart Traffic Light System by using Artificial Intelligence. S.S. Zia , M. Naseem, I. Mala , M.Tahir, T.Mughal,T. Mubeen(2019) Cuts traffic by reducing cars by 55% and wait time by 65% High cost, data reliance , sensor issues and limited scalability. 6. Traffic Management System. Lakshay Sharma,Pratyush Bhangalia(2023) Uses real-time data and AI to adjust signals, boost safety and cut congestion. Low testing, Scaling , privacy and emergency handling. 7. AI Traffic Control for Emergency Vehicles using Density Detection. Priyanka Abhang , Vinit Agrharkar , Kaushik , Piyush Mishra(2020) Uses video to ease traffic and speed up emergency response. Infrastructure , real-time load, emergency conflicts , weather. 8. Dynamic Traffic Light manage- Akash Gaur,Ayush Mavi,Bhumika Uses real time data to cut Camera accuracy, scaling , security &
- 7. PROBLEM DESCRIPTION ISSUES WITH PREPLANNED TRAFFIC SIGNALS Fixed Timing System: Green Signals operate on a fixed Schedule ,causing delays even when traffic is low. Inefficient Traffic flow : Vehicles must wait longer than necessary ,leading to congestion and fuel wastage. LIMITATIONS OF INDUCTOR-BASED SYSTEMS Detection Issues: Inductive sensors fail to detect small vehicles like scooters and bikes. Accuracy Problem :If a vehicle stops outside the sensor range ,it remains undetected. Not Suitable for India : Majority of Indian vehicles are small cars and two-wheelers making this system ineffective.
- 8. OBJECTIVE • Reduce Congestion-Adaptive signal control to minimize traffic delays • Improve Safety-Efficient traffic flow reduces accidents and emergency responses. • Environmental Benefits-lower fuel consumption and emissions by reducing idle time. • Improve the overall efficiency-Real time data analysis for optimized traffic movement.
- 9. PROPOSED METHODOLOGY • User Input Collection • Vehicle Detection and Counting - Utilize computer vision techniques(eg.,YOLOv8) for real-time vehicle detection. • Emergency Vehicle Detection - Create a separate model for detecting emergency vehicles using image processing techniques. • System Integration - Design a modular architecture for vehicle detection , emergency vehicle prioritization , and traffic signal control. • Testing - Conduct extensive field tests to evaluate system performance in real-world scenarios.
- 10. WORKFLOW DIAGRAM
- 11. RESULT
- 14. FUTURE SCOPE • Smart city applications to improve overall traffic flow and reduce congestion. • Improved accuracy through training on larger and more diverse datasets. • Edge computing deployment on traffic cameras for reduced latency. • Cloud integration for scalable processing and storage. • Collaboration with law enforcement for traffic violation monitoring
- 15. CONCLUSION The system encapsulates vehicle’s information with the use of object detection algorithms. It helps in understanding traffic management systems and drastically condenses the transit delay that is prevailing in the cities. It detects and counts the number of emergency vehicles that are approaching the signal in real time and alters the signal time to ensure effective congestion handling. The experimental results show that the parameters taken into consideration are suitable for real-time applications as they are precise and fast responsive. We had discussed a number of cases for handling collision which might arise in our system.
- 16. REFERENCE https://ieeexplore.ieee.org/Xplore/home.jsp https://en.wikipedia.org/wiki/ https://g.co/kgs/9Wems8Z https://ieeexplore.ieee.org/Xplore/home.jsp https://ijercse.com/article/dynamic-traffic-light.pdf https://ieeexplore.ieee.org/Xplore/home.jsp http://www.ir.juit.ac.in:8080/jspui/bitstream/123456789/10224/1/Traffi c%20Management%20System%20by%20Lakshay%20Sharma.pdf https://ieeexplore.ieee.org/Xplore/home.jsp
- 17. THANK YOU !