Convolutional Neural Network based Working Model of Self Driving Car - a Study
P. G. Chaitra, V S Deepthi, S. Gautami, H. M. Suraj, Naveen Kumar
Abstract
A self-driving car is a vehicle that senses its environment and navigates without human intervention and is a high research topic in computer vision that involves various subtopics and need to be deeply reviewed. To accomplish this, our paper discusses hardware and software components of a self driving car that includes usage of technologies such as Deep learning techniques namely Convolution Neural Networks, YOLO algorithm, Hough Transform Algorithms, Transfer Learning, Canny Edge Detection algorithm. Software components such as Arduino IDE, Raspberry Pi Cam Interface, Open CV, Tensor Flow, Carla simulators and hardware components such as Raspberry Pi 3, Arduino UNO, Pi Camera, sensors like radar, lidar are used to build a prototype of a self-driving car. This paper directs some of the complications in the existing technology and provides a few solutions that can be taken to overcome.