Machine Learning-Based Framework for Cybersecurity of Robotic System

Robotic systems are increasingly integrated into sectors ranging from manufacturing to health care, presenting significant cybersecurity challenges. Ensuring the security and integrity of these systems is critical to prevent malicious attacks that could compromise functionality and safety. This study introduces a machine learning-based framework designed to enhance the cybersecurity of industrial robotic systems. Leveraging the CICIDS 2017 dataset, which provides a comprehensive collection of network traffic data representing modern cyber threats, the framework demonstrates high efficacy in detecting and mitigating such threats. Experimental results show that the K-Nearest Neighbors and Decision Tree algorithms achieve accuracy rates of 99.01% and 99.83%, respectively, underscoring the potential of these techniques for real-time intrusion detection. These findings highlight the framework's effectiveness in protecting robotic systems against cyber-attacks, thereby fortifying their security in various application domains. The findings highlight the framework effectiveness in protecting robotic systems against cyber-attacks, thereby fortifying their security in various application domains.