Scholarlink

<p>In autonomous vehicles, cloud computing plays a major task in the development and operation. Meanwhile, conventional cloud computing methods face several challenges such as data storage, scalability, and cost efficiency in autonomous driving systems. To overcome these complexities, this study proposes the Modified Ensemble Weighted Support vector-based Adaptive Crossover Black Widow algorithm for cloud computing in autonomous vehicles. This study employs a Support Vector Machine for classification tasks and a Random Forest for preprocessing and feature selection. Also, this study utilizes an ensemble approach to enhance the classification performance and AdaBoost algorithm to enable real-time object detection. This real-time detection capability is important for ensuring the safety and efficiency of autonomous vehicles in various environments. Additionally, this study employs Gradient boosting to enhance the reliability and Adaptive crossover-based Black Widow Optimization to increase the efficiency of the proposed method. Validation of this method is conducted using the datasets namely the Karlsruhe Institute of Technology and Toyota Technological Institute dataset and Common Objects in Context 2017 dataset. The results demonstrate the proposed method achieved high efficiency with specificity, accuracy, precision, recall, and F1-score of 97.82%, 98.63%, 97.93%, 97.84%, and 97.85% respectively. Also, this study ensures more effective cloud computing solutions in autonomous vehicle systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing autonomous vehicle performance with ensemble weighted support vector-based optimization in cloud

  • 期刊论文

  • 发表日期：

    2025年1月21日

查看全文
  • Sethuraman Ravikumar,
  • Kiruthika Venkataramani,
  • Malu Manickam Yamuna Devi,
  • Surendran Subbaraj

摘要

In autonomous vehicles, cloud computing plays a major task in the development and operation. Meanwhile, conventional cloud computing methods face several challenges such as data storage, scalability, and cost efficiency in autonomous driving systems. To overcome these complexities, this study proposes the Modified Ensemble Weighted Support vector-based Adaptive Crossover Black Widow algorithm for cloud computing in autonomous vehicles. This study employs a Support Vector Machine for classification tasks and a Random Forest for preprocessing and feature selection. Also, this study utilizes an ensemble approach to enhance the classification performance and AdaBoost algorithm to enable real-time object detection. This real-time detection capability is important for ensuring the safety and efficiency of autonomous vehicles in various environments. Additionally, this study employs Gradient boosting to enhance the reliability and Adaptive crossover-based Black Widow Optimization to increase the efficiency of the proposed method. Validation of this method is conducted using the datasets namely the Karlsruhe Institute of Technology and Toyota Technological Institute dataset and Common Objects in Context 2017 dataset. The results demonstrate the proposed method achieved high efficiency with specificity, accuracy, precision, recall, and F1-score of 97.82%, 98.63%, 97.93%, 97.84%, and 97.85% respectively. Also, this study ensures more effective cloud computing solutions in autonomous vehicle systems.