ARGUS: An Autonomous Robotic Guard System for Uncovering Security Threats in Cyber-Physical Environments

While traditional surveillance and security methods remain widely used, they struggle to keep up with emerging threats. These systems lack autonomy, cannot process multimodal data in real time and are largely dependent on human intervention. This highlights the clear need for integrated solutions that combine autonomous robotic mobility, intelligent perception, and real-time contextual awareness, enabling faster and more effective responses to evolving security challenges. This paper presents ARGUS (Autonomous Robotic Guard System) an autonomous robotic platform for patrolling, detection, and response within cyber-physical environments. Unlike traditional systems, ARGUS operates independently, navigating, analyzing, and intervening in dynamic and potentially hazardous scenarios. The platform integrates advanced computer vision modules (facial recognition, vehicle, and bladed weapon detection), artificial intelligence for abnormal behavior assessment and contextual acoustic analysis, alongside embedded cyber protection capabilities through native Intrusion Detection Systems – IDS (Snort, Suricata). Additionally, ARGUS incorporates adaptive planning and routing algorithms (A*, D* Lite), Simultaneous Localization and Mapping (SLAM), and robust control strategies (backstepping, sliding-mode), enabling effective operation even in unstable, crowded, or unstructured environments. Beyond its autonomous functionalities, ARGUS can seamlessly integrate into a multi-agent network, coordinating real-time security responses with other mobile units and human operators. This paper details the system architecture, software and hardware components, implementation methodology, and testing scenarios, highlighting the advantages of adopting integrated robotic approaches for modern security applications.