Semi-Autonomous Vehicle Aadesh S. Tawte 1 , Tanmay N. Shindolkar 2 , Deep P. Maru 3 , Prof. Swati H. Shinde 4 123 Students, Department of Electronics and Telecommunication Engineering, K. J. Somaiya Institute of Engineering and Information Technology, Mumbai, Maharashtra, India. 4 Assistant Professor, Department of Electronics and Telecommunication Engineering, K. J. Somaiya Institute of Engineering and Information Technology, Mumbai, Maharashtra, India. I. INTRODUCTION In the recent years, there has been a major attention of research for autonomous vehicles in automotive engineering. One of the enormous technical challenges of autonomous vehicles is designing of dynamic path tracking, as a key component of the control system. The number of severe road accidents are increasing every year from 347 in 2020 to 389 in 2021. As a concern there is a need to automate the vehicle industry. The control method of trending autonomous vehicles which are based upon steering angle results in inaccurate tracking with respect to position onto the desired path, because as a vehicle moves in real-time, the steering wheel direction changes as per the way. Autonomous vehicle defines itself as a vehicle having extraordinary features that allows it to steer, brake and, accelerate with limited or no driver interaction. There are mainly two types of autonomous vehicles: semi- autonomous and fully-autonomous. Semi-autonomous cars are able to accelerate, brake and steer, and keep the distance from the car in front and also keep the lane at the speed of up to 130 km/h, but the need of driver is still required and is still in full command. Without any driver interaction, a fully autonomous vehicle is able to drive from one point to another. Recent research has proposed to construct the intelligent transportation systems, variable smarter suspensions, steering systems, torque distribution, steering by wire, and vehicle dynamic modelling improvement on designing more safer and intelligent vehicles. Varieties of studies are conducted about social impacts, regulations, human machine interfaces, and implementation methods of autonomous. The objective of Semi-Autonomous vehicle is to allow an algorithm to detect or learn various parts of a car such as wheels or other components of a car, objects/persons present around the self-automated car and as it has been exposed to much inputs, connections that has begun to develop to the endmost outputs and the steps to be provoked in the car in response. II. LITERATURE SURVEY Talking about autonomous vehicles (AVs) is a topic that has been discussed for several decades, but in the last decade the advances in computation, sensors and other hardware have allowed the development of this type of technology to be more real. All over the world, the evolution of self-automated vehicles has been observed, in cities such as Munich, Beijing, Paris, London, Boston, Singapore, San Francisco, Pheonix and New york, in companies such as Waymo, NuTonomy, Uber, Lift, Aurora, VW, BMW and others, just to mention a few. Driverless cars today sense their surroundings using radar, GPS, and computer vision, which could have an excellent margin for error, and the sensory information needs to be processed to navigate in pathways. A. Levels of Automation The Society of Automotive Engineers (SAE) stablished that there are six levels of vehicle autonomy: International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072 © 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 2924 2. Level 1 (Driver Assistance): The human driver has full control, but the vehicle helps with one or more driving functions, e.g., electronic stability control (ESC) and assisted braking. 3. Level 2 (Partial Automation): The human driver has primary control over the vehicle, but the vehicle can take 1. Level 0 (No Automation): The driver has full control and always performs all driving functions. ---------------------------------------------------------------------***-------------------------------------------------------------------- Abstract - In the fashionable era, the vehicles are centre to be automated to grant human driver relaxed driving within the numerous aspects are thought of that makes a vehicle machine-driven. Google, the most important network has started working on the self-driving cars since 2010 and still developing new changes to grant an entire new level to the automated vehicles. The major causes of accidents on the national highways are due to vehicle design and condition, speeding, driving on the wrong side, road engineering, use of mobile phones, jumping the red lights, etc. The need of semi- autonomous vehicle is important which is significantly automated and thereby reduces the death ratio due to accidents. Keywords – Raspberry Pi, Arduino UNO, Open Computer Vision, Machine Learning, Image Processing, Autonomous Vehicle.