Building a Self-Driving Vehicle in GTA-V Using Deep Learning and Convolutional Neural Network

Alzaib Karovalia
17 min readSep 17, 2020

GitHub: https://github.com/Alzaib/Autonomous-Self-Driving-Car-GTA-5

Introduction and Project Background

Artificial intelligence is now widely being used to perform several tasks that were previously performed by a human. An artificially created network can be trained to autonomously drive a vehicle, without any human intervention, creating a safer driving environment. These vehicles are equipped with different hardware such as camera, radar, and LIDAR, to collect the data in real-time. The data is then passed into complex neural networks to predict outcomes and control the car. Such networks can be built, train, and tested in a 3D realistic environment, without putting anyone on the risk. A simple self-driving vehicle can be built in a video game, without any need for such hardware. High-quality graphics, different weather conditions, unexpected behavior of nearby vehicles, and different traffic conditions are few of the many reasons that make Grand Theft Auto-V, an excellent choice to build and simulate a self-driving vehicle, as implemented by Sentdex [1].

--

--