Deep NN-Based Autonomous Vehicle Framework for Route Selection and Traffic Control Using Golden Levy Optimization

Nowadays, delivering autonomous systems with efficient quality as well as the amount of information from the preferred task is considered a challenging one, in which autonomous vehicles should possess a consistent vision of their workplace to achieve the driving operations robustly. In this research, a Golden Levy optimization-based Deep Neural Network (GLO-based DNN) is utilized for selecting optimized routes as well as controlling the traffic by assisting autonomous driving vehicles. The efficient path will be selected using the attained information from RSU through the optimized route selection. The traffic sign is controlled and modified with the traffic controller through the optimized DNN classifier. The convergence rate of the classifier is improved and minimizes the consumption time with the hybrid characteristics of Golden Levy optimization. Therefore, the attained value of QMA, SDTT, and throughput for the developed GLO-based DNN is 18.66 kmph, 5.42 s, and 12.89 % while considering 75 vehicles.