This is project implements basic self-driving techniques used in cars today, on a remote-control car. Here we use deep learning algorithm to train and implement a model that can navigate a remote-controlled car on a given track. Thus, this project is a good starting point for implementing neural network algorithm and get started with self-driving cars. This code draws heavy inspiration from the open source project called DonkeyCar. This open source project makes it easy to get started and then later add complexities to it. Some of the major hardware components used are mentioned below, while a complete list is provided in the project report.
• Exceed Magnet remote control car (HSP 94186 ), that we will automate. • RaspberryPi 3B+, as the central brain of this setup. • Servo Driver PCA 9685, controls the throttle and steering of the RC car. • Wide Angle Raspberry Pi Camera, the eye of the self-driving RC car.
Please refer to the project report for the background and results of the project.