RARRoad

Project Title

RARRoad

Project Area of Specialization Artificial IntelligenceProject Summary

RAR is an AI/computer vision based system that will provide assistance to the driver as road awareness is added in the rover using deep learning algorithms implemented on Raspberry Pi.

The system consists of two stages. The first stage focuses on  “Traffic sign board detection” and the second stage focuses on “Traffic sign classification and recognition”. Video input from the raspberry pi camera module will be taken from the surroundings and whenever a sign appears, the system will:

1) generate alerts to notify the driver through an Android application or

2) change the rover’s behavior accordingly

Project Objectives

1. Create an AI/computer vision based system that enables traffic sign detection and recognition in real time.

2. Create an “Alert Generation System” that will alert and notify the drivers whenever a traffic sign appears.

3. Create an IOT based system that will change the rover behavior according to the sign caught.

Project Implementation Method

The main purpose of this project was to introduce a solution for roslad safety problems. Traffic signs will be detected and recognized for drivers safety and prevent road hazards. The detection and recognition is being done with the help of YOLOv3, which is a fast deep learning algorithm based on CNN(Convolutional Neural Network).

Once the sign is detected and recognized through the created model which will be implemented on raspberry pi, the raspberry pi will communicate with an Android application on the user's phone and send alerts regarding the detected signs. In this way the driver will be able to know about the signs they missed due to human error and can avoid accidents. Another main part of the project is to change the rover's behavior on which this system is being implemented. When a sign is detected, raspberry pi will communicate it with the arduino present on the rover, and the rover will act according to the sign instructions.

The purpose of implementing this on the raspberry pi chip instead of cloud are connectivity issues. If the service is not available due to network interruptions, then that will also cause confusion and frustration on the driver's side, leading to accidents.

Benefits of the Project

• Assistance in Road Awareness/Driver assistance

  • Detect and Recognise the traffic signs and Alert the driver.

• Enforcement of Traffic Rules and Regulations

• Driver safety

• Reduction in road accidents

• Reduction in traffic congestion

Technical Details of Final Deliverable

The technical details after the final deliverable are as follows:

• Raspberry pi 4 with attached raspberry pi camera with the camera module on it. The camera will take videos of the surroundings to process them for the signs to be detected. This will help in realtime detection and recognition of traffic signs. Moreover, raspberry pi is being used as a processor as it is a cheap chip and is portable, meaning it can be attached on the rover. Further more, this will help in local processing excluding the need of network connections.
• Arduino UNO Rover is being used as a proof of concept to show the implementation of the project and controlling the behavior in response to the detected signs.
• An Arduino Uno R3 chip is used to control the motors of the rover. This will help in the black line detection implemented in arduino for the rover's autonomous movement. The rover can turn sideways by following the black line present on the road.
• Initially 4 signs are being detected by the Neural Network model to show the proof of concept.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther IndustriesCore Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT)Sustainable Development Goals Good Health and Well-Being for People, Industry, Innovation and Infrastructure, Sustainable Cities and CommunitiesRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	13280
Raspberry Pi	Equipment	1	9500	9500
Hdmi adapter	Equipment	1	400	400
Hdmi to vga converter	Equipment	1	150	150
Arduino UNO R3	Equipment	1	780	780
Arduino UNO Rover	Equipment	1	1100	1100
SD card	Equipment	1	480	480
IR obstacle avoidance senseors	Equipment	4	100	400
L298n motor driver	Equipment	1	250	250
18650 lion battery	Equipment	2	85	170
18659 2 cell battery holder	Equipment	1	50	50