Autonomous Vehicle Performance Enhancement via Ensemble Weighted Support Vector Optimization

Cloud computing plays a vital role in the development and operation of autonomous vehicles. Traditional methods for cloud computing in autonomous vehicles face several complexities, such as data storage, scalability, and cost efficiency. To overcome these challenges, a novel method named the Ensemble Weighted Support Vector Optimization (EWSVO) algorithm is proposed for cloud computing in autonomous vehicles. In this study, an ensemble based support vector machine (SVM) is utilized to improve the classification tasks performed by autonomous vehicles, and a Random Forest model is employed to assist in preprocessing and feature selection to enhance the efficiency of cloud computing. Also, the AdaBoost algorithm is implemented for object detection tasks such as identifying vehicles and obstacles in real-time cameras and the Gradient boosting algorithm is utilized to enhance the accuracy of the EWSVO method in vehicle detection. In this work, Adaptive cross-over-based Black Widow Optimization is employed to fine-tune the parameters and improve the efficiency of the proposed method. The study is validated on diverse datasets, including the KITTI and COCO2017 datasets, with 98.63% accuracy.