AI-Enabled Robotic Surveillance System for Critical Infrastructure
Manesh Thankappan, Neetha K Nataraj, K S Divya, Akshaya Jayaraj, Alen Joby, Abhilash Asokan · 2025
Traditional monitoring approaches often experience blind spots and high resource utilization at such a crucial time of security. In our study, we design a small-scaled autonomous vehicle model with smart sensors that is to be equipped with an AI model in a Raspberry Pi for effective and reliable self-sustained monitoring of critical infrastructure to overcome the problems. The core AI model was specifically designed to detect in real-time anomalies such as a fire, a fight, vandalism, explosion, etc. The AI model is a modified version of an open source repository available on github named “Real-world Anomaly Detection in Surveillance Videos (pytorch) [11]”. This technology ensures speedy response time by stopping the vehicle, starting video recording, and forwarding the real-time footage to security personnel upon detecting an anomaly. This innovative solution allows for mobilization and cost-effectiveness, improving surveillance coverage, and diminished dependence on constant human oversight. Exploiting advances in artificial intelligence and robotics, the project delivers a scalable real-time monitoring system deployable into a wide range of critical infrastructure environments. This initiative represents practical application of AI towards improving public safety and building a stronger security posture in facilities that house critical infrastructure.