Self Driving Car Using Machine Learning Implementation

  • S P Maniraj, Deep Chaudhary, Vankayala Hari Deep, Vishesh Pratap Singh, Anant Bhardwaj

Abstract

Self-driving vehicles or autonomous vehicles were being implemented by the use of AI and machine learning tool like computer vision for executing this process it is needed to use sensory inputs to train the AI to make decisions the sensory input consists of cameras, sonar, laser, radars like tools to give the input which is then processed by the model by using object detection which has two aspects that is image classification and image localization which is used to detect the object type and its distance from the observer point of view. Lane detection is the key property of the model which is used to keep the vehicles in the same lane, it depends on the combination of methods like line selection, hough transform, and spatial CNN.
All these features are necessary to train the algorithm and it also requires adaptive parameter tuning according to different times of the day and makes it versatile for different conditions that’s why adaptive features are necessary. This model is implemented using python 3 libraries like OpenCV to get a good accuracy for the model.

Published
2020-04-23
How to Cite
S P Maniraj, Deep Chaudhary, Vankayala Hari Deep, Vishesh Pratap Singh, Anant Bhardwaj. (2020). Self Driving Car Using Machine Learning Implementation. International Journal of Control and Automation, 13(02), 757 - 763. Retrieved from https://sersc.org/journals/index.php/IJCA/article/view/11222
Section
Articles