Self-driving car using artificial intelligence

Abstract

Self-driving autonomous vehicles are the solution for enhancing mobility intelligence related to driving. This project presents effective ways for implementation of a self-driving car. Proposed work is based on Artificial Intelligence, Computer Vision and Neural Networks. The proposed technology is implemented on a mini-robot car that was built from scratch, which uses Raspberry Pi and a camera as its core. The system built runs a script for complex task handling and sending appropriate commands to the vehicle. Image Processing techniques are also issued in the proposed system to identify various objects and traffic lights on the way. The system learns to autonomously navigate through reinforcement learning. Tensor Board is used to keep track of the working and efficiency of the trained Neural Network. The efficiency of the system is recorded at 96% as of now.