Skip to main content

Self-Driving Car using Artificial Intelligence: A Comprehensive Review

Page 1

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 11 Issue: 04 | Apr 2024

p-ISSN: 2395-0072

www.irjet.net

Self-Driving Car using Artificial Intelligence: A Comprehensive Review Devansh Singh1, Yash Sharma2, Roop Singh3 1 Student, Dept of CSE-AI, NIET Greater Noida, Uttar Pradesh, India,

2 Student, Dept of CSE-AI, NIET Greater Noida, Uttar Pradesh, India,

3 Professor, Dept. of CSE-AI, NIET Greater Noida, Uttar Pradesh, India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Self-driving cars powered by AI face a range of

transportation as we navigate this fascinating topic as seen in Fig.1 below.

challenges. Ensuring safety and reliability in all conditions is paramount, requiring robust decision-making capabilities in complex situations. Regulatory and legal hurdles vary widely between jurisdictions, while concerns over data privacy and security must be addressed to safeguard sensitive information and prevent cyberattacks. Ethical dilemmas, such as prioritizing passenger safety versus pedestrian well-being, pose complex moral questions. Building public trust and acceptance, interfacing with legacy infrastructure, and minimizing environmental impact are critical considerations. Using cutting-edge technology like Python, PyTorch, TensorFlow, and Unity's ML-Agents, this project delves into the fascinating realm of self-driving cars. We examine the underlying principles of autonomous cars, their AI-powered powers, and the transportation possibilities they present in the future. The project's goal is to analyse several self-driving car models that can navigate a simulation environment built in a safe and effective manner. We work to advance AI in autonomous driving by integrating components of perception, decision-making, and control. The reason for the review paper is that the existing review papers don't summarize all the types of techniques in a single paper; therefore, this paper represents a comprehensive review of self-driving cars using artificial intelligence.

Fig.1. Components of Self Driving Car Alheeti et al. [3] presented Autonomous vehicles, or selfdriving cars, mark a revolutionary paradigm change. Kang et al. [4] delivered that In the car industry, these vehicles are equipped with advanced sensors, AI algorithms, and a plethora of complex technologies that enable them to sense their environment, make judgments in real time, and drive themselves over roads.

Key Words: Traditional Driving, Self Driving Car, Radar, Sensors, Unity, Real time, Cameras.

1.INTRODUCTION

Soares et al. [5] presented Science fiction fans have been fascinated with the concept of driverless cars for decades, but new developments in artificial intelligence and machine learning have made this goal more attainable than in the past Soares et al.

Traditional driving refers to manual operation of a vehicle by a human driver, using physical controls like steering wheels and pedals, as opposed to autonomous driving where vehicles operate without direct human input. Agnihotri et al. [1] proposed Self-driving vehicles are a bright example of human creativity and innovation in the rapidly developing fields of artificial intelligence and technology. This project uses modern technologies like TensorFlow, PyTorch, Python, and ML-Agents in the Unity game engine to explore the world of autonomous automobiles.

The convergence of Python, PyTorch, TensorFlow, and MLAgents within the Unity environment is the central idea of this project. Mihalea et al. [6] developed the core programming language used to create the AI algorithms required for self-driving cars is Python, which is also very adaptable Mihalea et al. Gandhi et al. [7] proposed two wellknown deep learning frameworks, PyTorch and TensorFlow, offer the tools and resources needed to construct and train complex neural networks, allowing the car to sense its environment and make judgments.

Kudaka et al. [2] introduce about the specifics of self-driving cars, investigate the underlying technology that allow them to function safely and effectively, and dive into the revolutionary potential they hold for the future of

© 2024, IRJET

|

Impact Factor value: 8.226

|

ISO 9001:2008 Certified Journal

|

Page 1802


Turn static files into dynamic content formats.

Create a flipbook
Self-Driving Car using Artificial Intelligence: A Comprehensive Review by IRJET Journal - Issuu