Robotic systems in internet of things: addressing security challenges through threat modeling and penetration testing

The increasing prevalence of robotic systems across diverse domains has undoubtedly delivered numerous advantages. However, this proliferation has also exposed these systems to potential security threats, with the ability to cause significant human and financial losses. In this study we propose a proactive approach to risk management using threat modeling techniques. This approach provides a systematic framework for risk assessment, facilitating the identification and prioritization of the most severe threats for possible mitigation. Specifically, we provide Attack-Defense Tree (ADT) based methodology, an extension of the conventional tree formalism which incorporates mitigation nodes representing applicable counter measures at different hierarchical levels. This enhancement enables a comprehensive and structured analysis of security risks. The proposed ADT-based approach is tailored to reinforce the security of the camera component within robotic systems operating on the Internet of Robotic Things (IoRT) environment. Additionally, penetration testing is conducted to empirically evaluate the vulnerability of robotic devices to flooding-based denial-of-service (DoS) attacks. Experimental validation using the AlphaBot platform demonstrates the feasibility of executing a successful DoS attack on its camera module. These findings underscore the urgent need for anticipatory security strategies in IoRT systems and highlight the critical importance of robust, preemptive threat modeling to ensure cyber-resilience in modern robotic infrastructures.