Behavior Cloning for a Self-Parking Car

Behavior Cloning for a Self-Parking Car

Mr. Milind Khairnar 1,Karan Dantwala2, Jayesh Nilkanth Hegishte3

1 Artificial Intelligence and Data Science Department, Shah & Anchor Kutchhi Engineering College

 2 Artificial Intelligence and Data Science Department, Shah & Anchor Kutchhi Engineering College

 3 Artificial Intelligence and Data Science Department, Shah & Anchor Kutchhi Engineering College

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

Abstract - In the context of self-driving cars, Behavior Cloning can be used to train a model to perform complex driving tasks, such as parking, by observing human drivers perform these tasks. In a self-parking car, Behavior Cloning and genetic algorithms can be used together to train a model to perform the parking task. The model can be trained on a dataset of human driving behavior, where the expert driver is demonstrating the parking task. First, it allows the model to learn from human drivers, who have a wealth of experience and knowledge about driving and parking. Second, it enables the model to learn from a large dataset of driving behavior, which can improve its generalization performance. Finally, the use of genetic algorithms can help to optimize the model's performance and ensure that it is well-suited to the specific task of parking. Despite these challenges, Behavior Cloning and genetic algorithms have the potential to be powerful tools for training self-parking cars. By combining the strengths of imitation learning and optimization, these techniques can help to create models that are both accurate and efficient, and that can improve the safety and convenience of self-driving cars.

Key Words:

Car, self-Parking, Genetic algorithm, Accident safety, road, Virtual environment.