Development of an Autonomous Vehicle System with Remote Control and Cybersecurity Features

The transformation of transportation technologies has evolved from mechanical innovations to intelligent and connected mobility solutions. One of the most prominent examples of this transformation is autonomous driving systems, which offer significant benefits in terms of safety, comfort, and efficiency. However, their reliance on digital components such as wireless communication and remote access also introduces serious cybersecurity risks. In particular, mobile applications developed for vehicle control create new attack surfaces on connected systems. This study aims to develop a prototype autonomous vehicle system equipped with integrated cybersecurity features and managed via a mobile application that operates over a wireless network. Various cyberattacks—such as Deauthentication, Man-in-the-Middle, unauthorized remote access, and False Data Injection—were simulated under conditions approximating real-world scenarios, and the resulting network traffic data was collected. This data was then analyzed using machine learning techniques to automate the detection of attacks. The findings demonstrate the effectiveness of various algorithms, such as Naive Bayes, k-Nearest Neighbors (kNN), Random Forest, and Multi-Layer Perceptron (MLP), in detecting different types of attacks. Notably, the kNN algorithm stood out as the most successful model, exhibiting superior performance across various metrics, including accuracy, F1 score, and test time. This study highlights that not only the driving capabilities of autonomous vehicles but also their cybersecurity infrastructures must be enhanced using artificial intelligence-based methods. The developed system offers a model architecture that encompasses both connected mobility solutions and security, providing a solid foundation for future autonomous transportation technologies.