BAAO: Bayesian and Adam optimizer for fault prediction in self-driving software systems using deep learning-based hyperparameter tuning

Deep learning (DL) is crucial for advancing autonomous driving systems. The Basic requirement for successful and robust prediction of fault in autonomous driving systems is the assurance that the parameters involved in the calculation are always helpful in minimizing error. The Bayesian And Adam Optimizer (BAAO) is a novel approach that is represented and introduced in this paper by integrating Bayesian optimization and Adam Optimizer for fault prediction using deep learning-based hyperparameter tuning based on diverse learning rates. BAAO optimizes and improves fault prediction accuracy by utilizing an ideal learning rate, minimizing mean squared error (MSE), and mean absolute error (MAE), and maximizing R-squared related with each other in training and validation datasets. This paper proposes an optimizing model using the learning rate to the specific requirements of fault prediction tasks and demonstrates the effectiveness of Bayesian optimization in automating hyperparameter change in self-driving cars. The experimental results provide significant improvements over existing models and establish BAAO as an optimal model for fault prediction in autonomous driving systems. The later section of the paper includes a comparative study between the existing approaches and the proposed model and proves the optimization of the proposed model.