Deep Residual Network for Autonomous Vehicles Obstacle Avoidance

With the advancement of artificial intelligence and machine learning, autonomous automobiles are emerging as a lucrative subject of study and a source of interest for car companies. This framework contains our research. The fundamental purpose of this research is to propose an obstacle avoidance strategy for self-driving vehicles. We are looking at developing a model for high-quality obstacle avoidance prediction for autonomous cars that is based on images generated by our virtual simulation platform and then utilized with a ResNet50 deep learning technique. The primary challenge for an autonomous car is to move without collision. For autonomous vehicle simulation research, the suggested technique is feasible, efficient, and trustworthy. The performance of the proposed design is then compared to that of current architectures. The experimental results suggest that the ResNet50 design outperforms the other approaches tested.