Miniature Self Driving Car (P-0025) (MFN 6690)

Abstract

It is believed that human error is the primary cause of vehicular accidents. The human driver makes a driving mistake once in about 88 miles. Traffic related issues like traffic congestion, failure to obey traffic signals and such like can also be reduced to great extent if all the vehicles on roads will be autonomous. Autonomous or highly aware cars can positively affect tens of thousands of people. Automotive and technological companies are trying to make these vehicle as autonomous as possible. From the last few years self-driving cars have been emerged as one ofthe most innovative technology in the field of artificial intelligence. Societal and customer benefits of autonomous car are numerous in connection to safety, convenience, gasoline usage and the like. In the light of all the aforementioned existing traffic related issues and perks of autonomous car we come up with idea of developing a self-driving car. First and most important feature that is implemented on this car is Road Tracking. Second feature that we are implementing is road sign detection. Third feature that we are implementing is Road signal detection. Last feature that we implemented is "Obstacle avoidance". This prototype of self-driving car is using Raspberry pi (microcontroller) as an input unit, Pi camera is used, which provides front view of the car through which Stop sign and Road signal are detected and car is following the road. Ultrasonic sensor is used for avoiding obstacle. The Processing unit is Laptop, where artificial neural network is created to make predictions for car to move.