HIAD Highly Intelligent Automated Driver (KEY IJP************256)

• Bharath Kishore D ,Ranesh.s,Karthikeyan.p

Abstract

Abstract As autonomous vehicle technology develops quickly, a major problem continues to be guaranteeing safe navigation in dynamic and complicated surroundings. This paper suggests a unique reinforcement learning method called Q-learning for obstacle avoidance in self-driving cars. The goal is to create a system that is efficient and adaptable that can identify the best course of action to take to avoid a variety of obstacles in real time. The vehicle may automatically learn and improve its navigation strategy thanks to the integration of incentive reinforcement mechanisms, state representation, action selection, and sensor data processing in the proposed framework. By means of comprehensive simulation and empirical experiments, we exhibit the efficacy and expandability of the suggested methodology in a range of demanding situations.