Smart Driving Assistance for foggy weathers Using Machine Learning

Project Title

Smart Driving Assistance for foggy weathers Using Machine Learning

Project Area of Specialization Electrical/Electronic EngineeringProject Summary

The main theme and idea behind this project are that In Pakistan every morning in the winter starts with the Fog and it's very difficult to drive the car in the fog. The number of incidents due to fog is increased day by day. A recent road incident happened a week ago near the entrance of Mirpur Azad Kashmir. The Purposed idea helps the drivers to drive in foggy weather and reduces road accidents.

An increasing number of vehicles are equipped with onboard cameras. As perception sensors, they scan the surrounding field and supply data to the Advanced Driver Assistance Systems (ADAS) which generates an environmental model by employing computer vision techniques. Camera-based ADAS support the driver with timely risk identification to prevent accidents, e.g. by realizing an adaptive cruise control (ACC), traffic sign recognition, blind-spot detection or even performing an emergency brake.

While imaging sensors are performing well in good weather conditions, their efficiency suffers under adverse environmental influences such as heavy rain, fog, snow, the glare of the sun, sandstorm, and occlusion caused by dirt or other objects covering the windscreen such as ice. As a consequence, a vision-based ADAS may produce information of poor quality and the environmental model also becomes faulty.

Driving a car in reduced visibility conditions can easily lead to severe accidents. The fog has a huge impact on the visual range. A system that is able to alert the driver about this risk and automatically controls head- and tail-lights or reduces the speed of the vehicle might save lives. Furthermore, future vehicles may drive autonomously without driver interaction in environments with reduced complexity such as highways without cross-roads. These vehicles will make use of on-board cameras and need to trust their sensor inputs or, in doubt, hand over the vehicle control to the driver. To address this security issue, an increasing number of research groups are focusing on self-diagnosis mechanisms for camera-based advanced driver assistance systems. The overall aim of such a self-diagnosis mechanism is to classify optical threats and to develop a basic algorithm to estimate the information quality of cameras in order to warn the assistance system of possible critical working conditions. As a part of such a self-diagnosis mechanism, this work presents a novel approach based on frequency analysis of image blocks using power spectrum slope (PSS) for detecting a reduced viewing range of the camera, in particular, caused by fog

Project Objectives

· This hardware helps us to drive the vehicles in the foggy weather.

·This hardware makes assist in real-time with the help of camera and sensor
· The purposed hardware helps to minimize road accidents in foggy weather.

Project Implementation Method

We design hardware according to the requirements of the driver for this purpose we used a two-way identification system camera and the Sonar Sensor. The whole working of the hardware is controlled with the help of the Python Open Cv and the Arduino Uno R3. The Hardware components used in the implementation of this hardware are listed below.

Benefits of the Project

In Pakistan, when the winter season comes, there are a lot of Fogg in Punjab and Kashmir, due to which we do not see clearly for the next ten feet and that is why there are road accidents take place when an accident takes place their many vehicles collide with each other due to not being able to see the vehicles, which causes financial loss but also increases the loss of life many times. 

So this way we can eliminate these accidents. We will install a system in the car that will keep the driver active. We will first install an HD camera in the car and their high definition camera through which the driver will see all the oncoming vehicles. We will be able to see the vehicles comfortably. Then we will use a sensor Ultrasonic sencer standby which will send a signal and tell the oncoming vehicle and tell how far our vehicle is from the oncoming vehicle as if the next vehicle is very close. It will alarm the driver through an alarming signal that is turn left or tirn right.

Or we will have a server motor standby which will brake immediately and stop the vehicle. If God willing the vehicle still goes and collides with the next vehicles then we will use collision sensor which immediatly send sms to 1122 for first add. With the help of this project we can minimize the road accident due to foggy wheather.

Technical Details of Final Deliverable

The basic objective of our design is to ascertain the distance position and speed of the obstacle set at some distance from the sensor. Ultrasonic sensor sends the ultrasonic wave in various ways by rotating with help of servo motors. This wave goes in air and gets reflected back subsequent to striking some object. This wave is again detected by the sensor and its qualities is analyzed and output is shown in screen indicating parameters, for example, distance and position of object. Arduino IDE is utilized to compose code and transfer coding in Arduino and causes us to detect position or angle of servo motor and it is communicated through the serial port alongside the covered distance of the nearest object in its way. Output of all of this working is shown in the software called processing, it will display the input/output and the range of the object [4]. Implementations of the sensors are done in such a way that ultra-sonic sensor is attached on top of the servo motor because it has to detect the object and its distance. Arduino (micro-controller) will control the ultra-sonic sensor and servo motor and also powered will be given to both of them through micro-controller.

This Purposed system helps the drivers in foggy weather and helps the both public and private transport drivers.

Final Deliverable of the Project HW/SW integrated systemCore Industry ManufacturingOther IndustriesCore Technology OthersOther TechnologiesSustainable Development Goals Affordable and Clean Energy, Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	54630
1. Arduino Uno R3	Equipment	2	2950	5900
2. I2c Module	Equipment	2	200	400
3. Servo Motor	Equipment	1	8000	8000
4. Ultra Sonic Sensor	Equipment	4	350	1400
5. HD Camera	Equipment	1	22000	22000
6. smart vehicles like cars	Equipment	2	5500	11000
7. Connecting wires	Equipment	20	60	1200
8. Capacitors	Equipment	6	180	1080
9. Buzzer	Equipment	2	250	500
10. power supply for arduino	Equipment	1	1700	1700
11. Bridge Rectifier	Equipment	5	190	950
12. Transistors	Equipment	10	50	500