The increasing aging population presents a growing concern regarding fall incidents, which can lead to severe injuries, loss of autonomy, and increased healthcare costs. To address this issue, we propose G-Mov, an IoT solution that integrates robotics and Computer Vision (CV) for real-time fall detection and monitoring of elderly individuals. The system leverages a mobile robotic device equipped with a camera and an accelerometer, combined with CV and image processing techniques to detect falls. By incorporating ROS 2, OpenCV, and MediaPipe, the system performs real-time human pose estimation, ensuring accurate identification of fall events. Furthermore, MQTT and TLS protocols were used for secure and fast data transmission to an IoT cloud platform for real-time alerts and remote monitoring. The proposed system was evaluated in a controlled environment to test the performance of all components. The evaluation included both individual component tests and integration tests, using a simulated home setting.