Adaptive Traffic Light Timer Control (ATLTC)

Project Title

Adaptive Traffic Light Timer Control (ATLTC)

Project Area of Specialization Electrical/Electronic EngineeringProject Summary

Conventional traffic light control systems are based on fixed time intervals of the traffic lights. These conventional fixed traffic light controllers have limitations and are less efficient because they use a hardware, which functions according to the program that lacks the flexibility of modification and adaptation on a real time basis. Thus, due to the fixed time intervals of green and red signals there is excess and unnecessary waiting time on roads and vehicles consume more fuel. This eventually adds up to the environmental pollution and creates several health issues among the people on road and residing nearby. Also, these conventional traffic light control systems do not have any provisions to provide any information on traffic densities on various roads, which leads to traffic congestions. Thus, to make traffic light controlling and traffic regulation more efficient, we exploit the emergence of new technique called as Adaptive Traffic Light Timer Control (ATLTC). 

In the past this project is accomplished by using sensors(Ultrasonic or pressure) sensor now we using cameras instead of senosr for detecting the object on the intersections. 

Project Objectives

The objective behind the project is to limit the stoppage time and regulate the traffic flow by mean of the introduction of deep learning at all major traffic signals.

The project aims at reducing the jams in order to reduce traffic congestion, optimize traffic flow and help pro-actively manage traffic conditions. The project aims to increase the accuracy of the traffic camera for detecting the number of vehicles and fix the time with respect to it. Facing the limitations and major shortcoming of existing traffic signal control system, relying on the wealth of traffic control interaction conditions and data, and developing a collaborative control system with high degree of refinement, precision, and better responsiveness and intelligence are the objective need and development direction of traffic control technology. It can provide the scientific support for the development of future road traffic control system and can widely be used in new generation traffic control systems. Also it can improve the road network efficiency to a great extent, reduce traffic operation cost, prevent and mitigate traffic congestion at the intersections. And reduce energy consumption and emission.

Project Implementation Method

we using camera for object detection at different intersections so we using the machine learning techniques(CNN) for object detection. The code will be implemented in phython and the raspberry pi will use for the image processing and also controls the timing algorithm. CNN is the best technique for controlling the images. 

Benefits of the Project

Traffic control is one of the most important technical means to regulate traffic flow, improve the congestion and even reduce emissions. Its progress and development have always been accompanied by development of information technology, computer technology and system science. The self-adaptive traffic control system can adjust the signal timing parameters in real time according to control target of the manager and arrival characteristics of traffic flow at low intersections. Compared with timing control and actuated control, the self-adaptive system can make better use of overall traffic capacity of road network and effectively improve the efficiency of road network traffic.

As currently at the intercepts the traffic light system is conventional means the LED lights   changes with the fix amount of time the lights are not depends upon the vehicles at the intersections.

  A simple approach to improving traffic signal performance to makes the traffic light adaptive. The rapid development in Artificial intelligence (AI) and Machine learning (ML) has been considered as primary impetus for new industry revolution. Also, AI and ML technologies are best among solving traffic management system and control problems since a traffic system is hybrid, complex and stochastic. Other approaches are installed different types of sensors ultrasonic sensor, pressure sensor, induction loops in the road and sense the vehicles the microcontroller will count the vehicles and adjust the time according to the traffic. These techniques are less accurate the best way is to go through the AI and machine learning techniques.

Technical Details of Final Deliverable

We are using the resbperry pi for the controlling the camera and timing algorithm so the cameras will connected to the resbperry pi. The camera will take image and then forwarded to the resbperry pi the device will sense the vehicles check whether its density is high, low or moderate and then timing algorithm in resbperry pi will be applied for example we have 15 vehicles at the intersection so its the high density state so the green light will operating for 30 sec(Maximum). if 5 vehicles is detected at the intersection so its low density state the green light will operates for the 10 to 11 secs. if there is no vehicle at the intersection the green light will not operates this intersection will stay at red light until detection of vehicles records.  

Final Deliverable of the Project Hardware SystemCore Industry TransportationOther Industries Others , Security Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Good Health and Well-Being for People, Climate ActionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	70300
raspberry pi 4	Equipment	1	37000	37000
camera	Equipment	4	8000	32000
led	Miscellaneous	12	50	600
hardboard	Miscellaneous	1	500	500
wires	Miscellaneous	2	100	200