Smart Surveillance and Lifeguard Rover with Face Recognition and IoT for Chemical Plants

This work aims to develop a rover for operations within a petrochemical plant environment. The rover is controlled remotely via the internet using a live video stream and the communications facilitated by the MQTT (Message Queuing Telemetry Transport) protocol. To enable video streaming, an USB camera is attached to a Raspberry Pi. From the live streaming, presence of any humans inside the plant is detected using the Haar-cascade face detection algorithm running on Raspberry Pi. The Telebit and Flask are employed for displaying the video captured by the rover on a webpage. Real-time control and sensor data visualization are achieved through a secure webpage hosted on a personal server, ensuring enhanced security measures. User authentication is implemented to ensure authorized access to the webpage. The webpage, hosted on the Google Cloud Platform (GCP) - Virtual Private Server (VPS), provides comprehensive control and monitoring capabilities for the rover within the petrochemical plant environment. This research addresses the need for an internet-controlled rover that can operate in hazardous environments, such as petrochemical plants, with the aim of enhancing safety and efficiency in industrial operations.